Discover Computer Vision Projects Efficiently

I am looking for an experienced AI / Computer Vision developer to build a tool that can automatically:
Track a specific subject in a video, even when surrounded by multiple similar-looking subjects


Highlight or spotlight the subject in the output video so viewers can easily follow the subject
📌 Core Requirements
The tool should:
Process an input MP4 video and detect that subject throughout


Handle situations where subjects look similar, move quickly, turn, or become partially obscured


Output a processed video with a visual indicator (box, glow, spotlight, or similar) that follows the subject in a subtle but noticeable way


Maintain high accuracy and consistent tracking throughout the video
Output must be usable in commercial video projects.

🎯 Deliverables
Working script, program, or lightweight UI that performs:


Frame-by-frame tracking


Automatic highlighting overlay


Video export


Optional (but preferred):


JSON or CSV file with frame-by-frame coordinates


Ability to adjust the type of highlight effect (box, glow, outline, spotlight)


Documentation:


Installation instructions


How to run the program


Explanation of core methods used


All source code provided, with ownership assigned to me under NDA.



🛠 Recommended Technical Stack (Flexible)
Candidates may use any combination of:
Python


OpenCV


dlib


PyTorch / TensorFlow


FFmpeg or MoviePy for video output


I am open to proposals for alternate approaches if they improve accuracy.

👍 Ideal Candidate
Experience with object tracking, and video processing


Prior work with video subject tracking or similar projects


Ability to explain their approach clearly


Strong portfolio or sample code demonstrating relevant skills


Familiarity with accuracy challenges such as look-alike subjects, fast movement, partial occlusion, and varied lighting



❓ Please Include These in Your Proposal
Your proposed technical approach


Past examples of computer vision work


Estimated timeline and cost


Any initial concerns or data needs


Shortlisted candidates will be asked to sign an NDA before receiving sample videos and images.

We are seeking an experienced developer to create an AI-driven system for generating pharmaceutical documents. This system should streamline the documentation process, ensuring compliance with industry standards while enhancing efficiency. Ideal candidates will have a solid understanding of AI technologies and experience in the pharmaceutical sector. Key responsibilities include designing algorithms, developing user-friendly interfaces, and conducting thorough testing. If you are passionate about leveraging AI to innovate in pharmaceuticals, we want to hear from you!

• Describe your recent experience with similar projects

## Upwork Project Brief: AI Rugby Video Analysis Prototype (Computer Vision)

### 1. Project Overview & Goal

We are looking for an experienced Computer Vision (CV) and Deep Learning engineer to develop a prototype AI model capable of performing detailed analysis on rugby match footage.

The core deliverable is a functional prototype that can process a **five-minute clip of a rugby match** and output structured analysis data according to the detailed use cases below.

### 2. Required Technical Expertise

* Expertise in Computer Vision and Deep Learning (CV/DL).
* Proficiency in Python and relevant frameworks (PyTorch, TensorFlow).
* Experience with Object Detection and Tracking models (e.g., YOLO, R-CNN, DeepSort, etc.).
* Experience in video processing and handling large video files.
* Strong understanding of creating models for action/event classification in sports (e.g., classifying a ruck, maul, or scrum).

### 3. Deliverables & Prototype Requirements

The contractor must deliver a working prototype script/application that meets the following:

1. **Input:** Accepts a standard video file (e.g., MP4) of a rugby match.
2. **Processing:** Processes a minimum of **5 minutes** of the video clip.
3. **Output (Structured Data):** Produces a structured data file (JSON or CSV) containing:
* Timestamped bounding box coordinates for all detected objects (players, ball).
* Player IDs/tracking information (maintaining the identity of each detected player across frames).
* Timestamped identification and classification of critical game events (Ruck, Maul, Scrum, Line out).
4. **Output (Visual):** A sample output video file with bounding boxes and event labels overlaid on the footage, demonstrating the model’s accuracy.
5. **Code & Documentation:** Clean, commented code and detailed documentation on installation, environment setup, and how to run the model on new video files.

### 4. Specific AI Model Capabilities (Use Cases)

The model must be designed to enable the following detailed use cases, which define the required analysis output:

| Category | Required Model Action |
| :--- | :--- |
| **Sports Analysis** | Identify and track player positions and movements throughout the game. **Automated identification and timestamping of critical game moments** (Ruck, Maul, Scrum, and Line out). |
| **Player Performance** | Monitor and evaluate individual player performance by tracking their involvement with the ball and participation in Ruck, Maul, Scrum, or Line out. |
| **Content Creation** | Track specific actions such as player movements or ball interactions for generating highlight content. |
| **Predictive Insights** | Provide structured data (coordinates, events) that can later be used to gather predictive insights for betting or fantasy leagues. |

### 5. Technical Considerations & Data Strategy

#### **A. Baseline Model (Roboflow Link)**

We have identified an existing open-source project that may serve as a useful baseline.

* **Reference:** `https://universe.roboflow.com/university-of-tokyo-eqewd/rugby-model`
* **Note:** This model and dataset are **open source** and should be evaluated. The contractor should assess if this foundation can be leveraged via transfer learning (e.g., using its player/ball detection) and then augment it with custom classification logic to meet the specific requirements (especially event classification: ruck, maul, scrum, line out).
* **Proposal:** Please specify in your proposal whether you plan to use this baseline, a different open-source model (like YOLOv8/v10), or build a custom architecture, and justify your choice.

#### **B. Data Requirements**

* The contractor must detail the strategy for obtaining and/or labeling data required for the event classification (Ruck, Maul, Scrum, Line out), as this data may not be fully available in the Roboflow dataset.

### 6. Budget and Timeline

This is a fixed-price project for the **prototype phase** only. Please provide a detailed breakdown of your estimated time and cost, including:

1. Model selection/setup and environment configuration.
2. Data preparation/labeling (if necessary).
3. Model training and refinement.
4. Integration, testing, and documentation.

Title
Computer Vision Engineer (Human Motion Analysis | Camera-Based AI)

Description
We are looking for an experienced Computer Vision / AI Engineer to collaborate on an early-stage project involving camera-based human motion analysis.

The work focuses on:

Detecting and tracking human body movement using a standard camera

Analyzing movement patterns and posture

Producing structured outputs suitable for real-time feedback systems

This is an MVP-stage project with potential for long-term collaboration if the initial phase goes well.

Requirements

Strong experience in Computer Vision

Hands-on experience with pose estimation / human movement tracking

Familiarity with tools such as OpenCV, MediaPipe, OpenPose, MoveNet, or similar

Experience with Python and TensorFlow / PyTorch

Ability to work with imperfect camera conditions (lighting, angles, occlusion)

Nice to Have

Experience in sports, fitness, rehabilitation, or motion analysis projects

Experience optimizing models for real-time or mobile usage

Project Type

Contract / Freelance

Duration: Short-term for MVP (with possible extension)

To Apply, Please Share

Relevant past projects (GitHub / demos welcome)

I am looking for someone to build a scoreboard OCR (Optical Character Recognition) program to add to a previously build graphics software.

The program should ask the user to select a video feed, allow the user to draw sections over the feed and the program should convert numbers/ letters in those sections into text form.

Skills: Windows Desktop, Software Architecture, OCR, Software Development, Image Processing, Computer Vision, Electron JS

Hello,
I’m looking for an experienced developer to build a semi-automated Etsy listing system that is fully compliant with Etsy policies.

The goal is to speed up listing creation, not to violate Etsy rules or bypass safeguards.

✅ CORE REQUIREMENTS

1. AI Image Generation Pipeline

Generate product images using AI (Midjourney / Stable Diffusion / DALL·E or similar)

Support prompt variation to avoid repetitive visuals

Images must look natural, realistic, and suitable for Etsy

No copyrighted characters or restricted content

2. AI Listing Content Generator

Automatically generate:

SEO-optimized title (Etsy best practices)

13 tags

Category & attributes

Product description:

SEO intro

Bullet points

Care / delivery info

Content must be unique per listing

No keyword stuffing or spammy behavior

3. Etsy Upload (SAFE & COMPLIANT)

Use Etsy official API only

Respect rate limits

No scraping, no browser automation (Selenium/Playwright)

Prefer manual approval step before publishing

4. Human-Like Workflow

Designed for:

1–2 Etsy shops

Low daily volume (1–5 listings/day)

NOT mass posting

NOT multi-account abuse

NOT automation meant to bypass Etsy systems

⚠️ IMPORTANT COMPLIANCE RULES

This project must strictly follow:

Etsy Seller Policy

Etsy API Terms of Use

No account creation automation

No IP rotation, fingerprint masking, or evasion logic

No fake shipping or misleading listings

If any feature risks account suspension, it should not be implemented.

🛠️ TECH PREFERENCES

Backend: Python or Node.js

Optional dashboard (simple web UI)

Clean, documented code

Modular structure (image → content → upload)

📦 DELIVERABLES

Working MVP

Setup instructions

Code documentation

Basic error handling & logs

⏱️ ESTIMATED SCOPE

MVP: 20–40 hours

Future improvements possible after testing

Please include in your proposal:

Similar automation or marketplace experience

Tech stack suggestion

Estimated timeline

Any compliance risks you foresee

-Umut

📍 Job Overview

We are seeking a highly organized, research-oriented Virtual Assistant to support academic and applied AI collaboration outreach. The role focuses on identifying ongoing AI research projects and papers in Computer Vision, Healthcare AI, NLP, and Machine Learning, profiling the founder’s technical background, and managing professional outreach and email communication with researchers, labs, and project leads.

This is not a generic administrative role. The ideal candidate understands AI research culture, academic collaboration norms, and professional communication in research environments.

⸻

🎯 Core Objective

The primary objective is to identify, initiate, and manage collaboration discussions for legitimate co-authorship and contribution opportunities in ongoing AI research projects, and to shortlist confirmed projects that align with the founder’s expertise and interests.

⸻

🧠 Key Responsibilities

1. Profile & Positioning
• Study and understand the founder’s:
• Academic background
• Technical skill set (ML, CV, Healthcare AI, NLP)
• Research interests and applied AI focus
• Convert this information into:
• A concise research collaborator profile
• Short and long versions for different outreach contexts
• Ensure positioning is ethical, accurate, and contribution-driven (no ghost authorship).

⸻

2. Research Opportunity Discovery
• Actively search for ongoing AI research projects and papers, including:
• Pre-submission works
• Workshops, special issues, and active lab projects
• Open collaboration calls
• Platforms to be used (but not limited to):
• Google Scholar
• ResearchGate
• Academia.edu
• AMiner
• Conference workshop pages (NeurIPS, MICCAI, CVPR, ICCV, EMNLP, AAAI, etc.)
• University lab websites
• Relevant mailing lists and research forums
• Focus domains:
• Computer Vision
• Medical / Healthcare AI
• NLP
• Multimodal AI
• Applied Machine Learning

⸻

3. Outreach & Email Management
• Draft high-quality, personalized outreach emails, including:
• Introduction
• Clear value proposition
• Specific contribution angles (experiments, modeling, evaluation, writing)
• Handle:
• Initial outreach
• Follow-ups
• Back-and-forth communication professionally
• Maintain professional academic tone (not sales-like, not generic).
• Adapt messaging depending on:
• Professor vs PhD student vs industry researcher
• Academic vs applied research context

⸻

4. Point-of-Contact & CRM-Style Tracking
• Maintain a structured tracker (Google Sheets / Notion / Excel) containing:
• Researcher name
• Institution / Lab
• Research topic
• Contact details
• Outreach status
• Response status
• Confirmed collaboration interest (Yes / No / Pending)
• Ensure clean documentation and traceability.

⸻

5. Shortlisting & Reporting
• Prepare weekly or bi-weekly reports summarizing:
• Outreach conducted
• Responses received
• Confirmed collaboration opportunities
• Shortlist validated, confirmed research projects with:
• Topic summary
• Expected contribution scope
• Timeline
• Target venue (journal / conference)
• Final deliverable: a curated list of real, active research projects the founder can choose from.

⸻

🧩 Required Skills & Background

Mandatory
• Strong written English (academic/professional level)
• Excellent research and information synthesis skills
• Ability to understand technical concepts at a high level
• Experience with email outreach and professional communication
• Strong organizational and documentation skills

Highly Preferred
• Background or exposure to:
• Machine Learning / AI
• Academic research or publishing
• Prior experience:
• Working with researchers, PhDs, or startups
• Academic virtual assistance or research coordination
• Familiarity with:
• Research collaboration etiquette
• Authorship ethics and contribution norms

⸻

⚠️ Ethics & Boundaries (Important)
• No ghost authorship
• No pay-to-publish schemes
• No unethical journal practices
• All collaborations must be contribution-based and legitimate
• Quality over quantity is expected

Pakit, Inc. is a startup company in the automation logistics space. Our goal is to develop a simple yet powerful automation (hardware/software) tool to help small business automate their LTL freight as soon as it hits the scale.
We are developing an MVP for a dimensional scanning camera that has the capability to extract written words, package dimensions, and blue toothed weight information. The information will be sent to a U.I. and freight carrier api's.

Features:
Computer vision compatible camera
OCR
API Integration
Bluetooth Compatibility
Wi-Fi Compatibility

Skills:
(Professional)
- Regular and detailed communication
- Solid Organization
- Consistent documentation
- Attention to detail

(Technology)
- Python & C
- MongoDB or similar
- Embedded systems/ PCB design
- A.I./ML integrations (preferred - not required)

If you have interest in A.I/ML, logistics, and helping small business gain a competitive edge in this corporate dominated economy reach out to set up a meeting!
www.pakitinc.com

• Describe your recent experience with similar projects
• Include a link to your GitHub profile and/or website
• What frameworks have you worked with?

Essential:
- Python with asyncio
- Selenium/Playwright automation
- Web scraping (anti-detection)
- Proxy rotation implementation
- Payment form automation
- WebSocket handling

Advanced:
- Computer vision for CAPTCHAs
- Browser fingerprint spoofing
- Reverse engineering WhatsApp
- Residential proxy networks
- Payment card tokenization

I need your help sourcing, screening, and presenting a seasoned AI/ML engineer who is comfortable communicating in fluent English. The ideal candidate must demonstrate strong, hands-on expertise in Natural Language Processing, Computer Vision, and core Machine Learning Algorithms. Please focus on professionals who can speak in depth about recent production work, architectural decisions, and measurable business impact—this is a senior-level search.

Here’s what I expect from you:
• A concise shortlist of qualified candidates, each accompanied by a brief skills matrix and salary expectations.
• Your notes from an initial English-language technical screen confirming depth in NLP, CV, and ML fundamentals.
• Coordination of interviews and smooth hand-off once we decide to proceed.

If you already maintain a network of AI specialists, even better. Let me know your sourcing strategy and typical turnaround so we can kick this off right away.

Skills: Machine Learning (ML), Computer Vision, AI Consulting, Natural Language Processing, AI Text-to-speech, AI Model Development, AI Research, AI Development

We are seeking a skilled AI developer to create a model capable of analyzing .JPG images to accurately count the number of livestock present. The ideal candidate will have experience in computer vision and machine learning, and will be able to implement an efficient solution. You will be responsible for data preprocessing, model training, and evaluation of accuracy. Strong problem-solving skills and attention to detail are essential for this project. If you have a passion for AI and agricultural technology, we would love to hear from you!

Relevant Skills:
- Machine Learning
- Computer Vision
- Python Programming
- TensorFlow or PyTorch
- Data Preprocessing
- Model Evaluation

We are seeking an experienced developer to create a 3D Spatial Vision AI system for continuous inventory counting. The ideal candidate will have expertise in computer vision, machine learning, and AI algorithms to design a system that accurately tracks and counts inventory in real-time. Your role will involve collaborating with our team to identify requirements, develop algorithms, and implement the solution. If you have a passion for innovative AI solutions and a proven track record in similar projects, we'd love to hear from you!

We are seeking a skilled developer with expertise in the VisionLib SDK to create a custom object tracking AR application for iPhone.

• Describe recent projects where VisionLib SDK was used.

We are seeking a skilled freelancer to develop a prototype utilizing computer vision and machine learning techniques. The ideal candidate will have experience in creating innovative solutions and be able to translate complex requirements into functional prototypes. You will work closely with our team to identify key features and deliver a working model that demonstrates the capabilities of your solution. Strong problem-solving skills and the ability to work independently are essential for this role.

• Describe your recent experience with similar projects
• How do you deal with outliers or missing values in a dataset?
• What tools do you use for data mining and visualization?
• What techniques would you use to clean a data set?

OCR
Computer vision
Hardware design
Api integration
Bluetooth compatibility
Wifi compatibility

Creation of a Dataset for Automatic Detection of Electrical Components in Residential Plans (YOLOv8 / Roboflow / Computer Vision)

ElectraPlan AI is an artificial intelligence system designed to analyze construction blueprints and technical documentation (in PDF or image format) to automatically generate a complete list of electrical materials required for residential or commercial buildings.

Skills: Data Processing, Engineering, CAD/CAM, PDF, AutoCAD, Computer Vision, Deep Learning, Adobe Acrobat, Data Management

I’m building a cross-platform, AI-powered editor that can take raw footage, understand what matters, and output a polished cut with zero manual effort. The core of the product is smart video editing, so the algorithms you design must automatically:

• trim silent or low-action sections,
• correct colour balance and exposure,
• apply tasteful filter effects that users can toggle on or off.

The finished application has to run natively on Windows, macOS, and Linux. I’m comfortable if you lean on Python, C++, or a mix (OpenCV, FFmpeg, TensorFlow / PyTorch are all fair game) so long as the final binaries are easy to install and the code is well-documented.

Deliverables
1. Source code with build scripts for the three desktop OSs.
2. A simple GUI (Qt, Electron, or similar) that lets users load a clip, press “Auto-Edit”, preview changes, and export.
3. Pre-trained models or clear training instructions for the trimming, colour, and filter components.
4. README showing installation, sample command line usage, and minimum system requirements.
5. Short demo video proving automated trimming, colour correction, and filter effects work on each platform.

Acceptance test: I will feed a ten-minute 1080p clip; the tool should return a coherent, colour-balanced, filtered 3-minute highlight reel in under 5 minutes on a mid-range laptop.

If you have prior work in intelligent editing or computer vision, please share a link or brief summary with your bid.

Skills: C Programming, Python, C++ Programming, ActionScript, Video Editing, Qt, OpenCV, Video Processing, Computer Vision, Deep Learning

We are looking for an experienced AI/ML Engineer to enhance our mobile-first product-recommendation web app that analyzes a user’s hair through image input, identifies key attributes, and recommends personalized products based on hair type and goals. The core system is already functional, and we now want to upgrade its accuracy, performance, and recommendation intelligence.

Responsibilities
- Improve and fine-tune the hair image analysis pipeline (classification, segmentation, texture/damage detection).
- Enhance the ML model accuracy and optimize inference for mobile-first performance.
- Refine our recommendation logic, incorporating user goals, hair attributes, and decision-layer improvements.

Required Skills
- Strong experience with computer vision, image classification/segmentation, and model optimization.
- Expertise in Python, TensorFlow/PyTorch, and modern ML workflows.
- Familiarity with recommendation systems, embeddings, and personalization models.
- Ability to integrate ML models into production web apps with efficient inference pipelines.

We need an expert for help us with training a Yolo based custom model. We have a setup for grading fruits (tomato and apple) using a camera feed connected to an RPI5 board. We want to consult with an expert who has worked with training/optimizing Yolo models for achieving reasonable framerate with RPI5.

We need an expert for help us with training a Yolo based custom model. We have a setup for grading fruits (tomato and apple) using a camera feed connected to an RPI5 board. We want to consult with an expert who has worked with training/optimizing Yolo models for achieving reasonable framerate with RPI5.

We are seeking an experienced AI / Computer Vision Engineer to join our team on an exciting project focused on developing advanced image recognition solutions. The ideal candidate will have a solid background in machine learning, deep learning, and computer vision techniques. Responsibilities include designing algorithms, implementing models, and optimizing performance. If you have a passion for AI and want to contribute to cutting-edge technology, we would love to hear from you!

I need a prompt engineer to refine and optimize prompts for extracting electrical symbols from blueprint PDF pages using OpenAI Vision.

Tasks:
• Build structured prompts (JSON output)
• Optimize symbol detection
• Add symbol disambiguation logic
• Help create prompt templates for different blueprint styles

Deliverables:
• Final prompt pack (5–10 variations)
• Accuracy improvement suggestions

Please share:
• Experience with vision models
• Examples of JSON extraction prompts

Embedded Engineer – Wearable Smart Glasses + Pocket AI Unit (Prototype)
Project Overview
We are looking for an experienced embedded systems engineer or small team to
design and build a working prototype of a wearable visual reading assistant,
composed of:
1. 2. Smart Glasses (Head Unit) – image capture + bone conduction audio
Pocket Processing Unit – AI processing, Bluetooth communication, power
management
The system must be silent, reliable, and discreet, suitable for everyday use.
This is a hardware + firmware project. Industrial-quality execution is required.
Scope of Work
A. Head Unit (Smart Glasses)
• Base frame: Carrera 8825 – Matte Black Acetate
• Camera:
o OV5640 (5MP), pinhole lens
o Mounted in bridge area
o Aperture ≤ 1.0 mm
o Optical insert using black PMMA (IR-pass acceptable)
o Matte surface finish, sanded or bead-blasted to match frame texture and
minimize reflections
o All LEDs disabled or removed
• Audio:
o Stereo bone conduction transducers integrated in temples
• Power:
o ~350 mAh LiPo battery
o Qi wireless charging (no exposed USB ports)
• Controls:
o Power via Hall sensor + magnet (hinge-based)
o No accelerometer
B. Pocket Unit
• Core: Raspberry Pi Zero 2 W
• Enclosure:
o 3D printed ABS or PETG
o Passive cooling only (no fans)
• Interface:
o Two recessed physical buttons (Volume + / −)
• Power:
o 5,000 mAh battery
o BMS
o USB-C charging
• Connectivity:
o Bluetooth Classic (SPP profile)
C. Communication & Firmware
• Bluetooth Classic only (Wi-Fi disabled on head unit)
• Image preprocessing on head unit:
o Grayscale conversion
o JPEG compression
• Target latency: < 5 seconds end-to-end
D. Software Architecture
• OCR + AI-based text understanding
• Cloud AI supported (API-based)
• No API keys hardcoded
• System must read from SD card:
o config.txt → system parameters
o prompt.txt → AI prompt text (user-editable)
• AI provider must be modular (Gemini / OpenAI / equivalent)
Deliverables
• Fully assembled hardware prototype (glasses + pocket unit)
• Firmware for head unit
• Software stack for pocket unit
• SD card configuration format documentation
• Power consumption report
• Assembly and maintenance documentation
Required Skills
• Embedded systems (ESP32 / SBC)
• Bluetooth Classic (SPP)
• Computer vision & OCR
• Low-power hardware design
• 3D printing & enclosure design
Nice to Have
• Wearables experience
• Bone conduction audio integration
• Previous camera miniaturization work
To Apply
Please include:
• Relevant past projects
• Which parts you would handle personally
• Estimated timeline
• Estimated budget range

Skills: Python, Product Design, Raspberry Pi, RESTful API, Soldering, Embedded Systems, PCB Design and Layout

We are looking for a developer to build a dual-purpose AI tool that connects visual data, technical documentation, and our internal database. The goal is to automate the analysis of user-uploaded images and allow users to query our internal records using natural language.

The Core Requirements

1. Visual Analysis & Reference Matching The system must handle an image upload workflow:

Image Recognition: The tool must analyze uploaded photos to identify specific objects, conditions, or anomalies.

Document Retrieval: Once an object or condition is identified, the system must search our established repository of reference documents (PDFs/manuals).

Output: The system must return the specific rules, codes, or text references from the documents that apply to what was found in the image.

2. Database Interaction (Natural Language Search) The system must serve as a conversational interface for our internal data:

Data Querying: Users must be able to ask plain-language questions about other users, historical logs, values, or entries.

Cross-Referencing: The AI needs to search across all structured data to retrieve accurate information without the user needing to write code or queries.

Deliverables

A secure chat interface that handles both image and text inputs.

A backend system that indexes our reference documents for fast retrieval.

Integration with our existing user database/records.

A solution that prioritizes data accuracy and cites its sources.

Ideal Skills

Experience with Computer Vision (identifying objects/attributes in images).

Experience with RAG (Retrieval-Augmented Generation) for searching document libraries.

Experience connecting AI agents to structured databases (SQL/NoSQL).

I can develop a smart system to monitor staff activity in your restaurant, mall, coffee shop, or retail

space. This solution provides actionable insights to optimize staff productivity, improve efficiency, and

support data-driven management decisions.

The system focuses entirely on staff monitoring, tracking activity levels, idle times, workstation usage,

and interactions among team members all while ensuring privacy and compliance.



How It Works

1. Camera Feed Integration

a. Connects directly to your existing CCTV/IP cameras via RTSP.

b. Supports live feeds or recorded footage for testing.



2. Staff Detection & Tracking

a. Uses YOLOv8 to detect staff in the video feed.

b. Tracks individuals across frames using Deep SORT, giving each staff member a unique ID.

c. Staff can be tagged by role (e.g., chef, waiter, supervisor) in the dashboard for analysis.



3. Behavior & Zone Monitoring

a. Define zones such as workstations, service counters, storage, or break areas.

b. System tracks time spent in each zone and logs whether staff are active or idle.

c. Generates alerts if a staff member is inactive for too long or spending excessive time in

non-work zones.



4. Dashboard & Analytics

a. Web-based interface shows daily and weekly summaries.

b. Timeline graphs, pie charts, and activity trails visualize staff performance.

c. Export reports in CSV or PDF for management review.



Use Cases

• Restaurants & Cafes: Monitor waiter and kitchen staff productivity, identify idle periods, optimize

shift allocation, and reduce operational inefficiencies.

• Retail Stores & Malls: Track staff presence in different zones and optimize coverage during peak

hours.

• Offices & Warehouses: Ensure employees are active in assigned work areas, measure

workstation efficiency, and support performance reviews.



Development Plans & Features

**All three planes are in each milestone**

to review the full details please review the attachements



Note:

The client has agreed to a discounted price of $6,000 for the initial Phase-1 (Startup Plan). After

reviewing the results and based on his satisfaction and feedback, he may choose to increase the

payment up to the original plan price of $8,500, but this is optional and fully his decision.

The current confirmed cost is $6,000.



Why Choose This System

• Quick setup with minimal configuration.

• Focused metrics on staff productivity and activity.

• Reduces inefficiencies, improves performance, and ensures accountability.

• Provides clear, actionable insights for management decisions.

• Designed to scale for multiple locations or high-traffic environments.

@everyone ┇ BIG REQUEST ┇ NO UPFRONT ┇ PAY ON FULL DELIVERY ONLY

Helios 2 + Titan Two │ COMPLETE Private Discord Dashboard + Full AI Suite + ALL Scripts From Scratch
NBA 2K26 + UNIVERSAL FPS (Destiny 2 · CoD · Apex · Battlefield 6 · Valorant · XDefiant + ANY future FPS)

Need ONE god-tier dev (or small duo) who can deliver the whole package 100% finished and working on every title listed.

Everything built from zero:
• Private Discord bot + clean Atlas/Phoenix-level dashboard (zero lag, beautiful embeds)
• ALL Titan Two GPC scripts written fresh (anti-recoil, rapid fire, perfect release combos, aim injection, macros – everything modular)
• Helios 2 running all CV/AI (YOLOv8/v9/TensorFlow best models)
• Beginner mode = braindead easy sliders & presets
Advanced mode = raw variables, script editor, custom model training

Discord Dashboard (looks & feels premium)
• Live sliders/buttons/dropdowns/graphs/tooltips
• Save/load/share profiles with thumbnails
• Real-time stats feed + visual overlay previews
• Password channels + auto model/script updater

NBA 2K26 – 100% Green Machine
• Meter + No-Meter perfect release (triple detection: meter/animation/audio)
• Every shot: jumpers, fades, hops, post spins, stepbacks, leaners, floaters, dunks
• Auto-detect contest/stamina/badges and adjust timing
• One-click retrain after every patch (guided for noobs, full PyTorch for pros)

UNIVERSAL FPS – Works FLAWLESS on ALL these games + any future title
• Destiny 2 (perfect tracking on sliding shotgun apes, bow/sniper swaps, super tracking)
• Call of Duty (Warzone/MW3/BO6 – bloom control, perfect movement tracking)
• Apex Legends (legend ability detection, ultimate tracking, zip-line prediction)
• Battlefield 6 (2042 & next BF – vehicle lock-on, massive player count handling)
• Valorant (agent outlines, spike timer sync)
• XDefiant + every new shooter that drops

Aim Features (all games)
• YOLO head/body detection running on Helios 2
• Rage / Legit / Silent / Triggerbot / Memory aim
• Humanized smoothing, random offset, bezier curves
• Live recoil tables per weapon (drag-drop for noobs, raw values for pros)
• FOV circle, priority (closest/head/low HP), bone selection
• Prediction for movement, sliding, jumping, supers (D2), ziplines (Apex)
• Anti-recoil + no-spread + quickscope + auto-fire on lock

Tech
• 100% local, encrypted, zero cloud
• All scripts + models delivered ready to flash and run
• Future-proof templates so new games just need a quick model swap

Payment
• ZERO upfront, zero milestones
• Full payment ONLY when I flash everything, join the Discord, and it works perfectly in-game on NBA 2K26 + Destiny 2 + CoD + Apex + Battlefield 6 (I’ll test all)

DM me with:
• Short clips/screenshots of your past Helios 2 + Titan Two full packages (especially multi-game FPS proof)
• How long until I get the complete ready-to-use version
• Your price (paid only on full delivery)

New talented devs welcome — if you’re actually good with CV + GPC and can make it noob-friendly but cracked for pros, slide in. Proof gets priority.

Let’s make the only setup anyone will ever need in 2026 and beyond.

Skills: PHP, JavaScript, Animation, After Effects, Game Development, Computer Vision, Discord API, AI Development

I’m a business owner based in Dubai running an operating company (projects, sites, reporting, follow-ups).

I’m looking for someone who has actually built AI automations for real businesses and can mentor me 1-on-1 while we implement them.

What I want to achieve :
Build 2–3 real AI automations that I will use in my company
Focus on LLMs + automation, not data science
Learn by doing, with clear outputs each month

Tools I want to work with :
ChatGPT / LLMs
Make.com (or Zapier if needed)
Google Sheets / Excel
Email / WhatsApp style workflows

Engagement structure :
2–3 live sessions per month (60–90 mins)
Clear homework between sessions
Budget: USD 15–25/hour
Starting with a paid 45–60 min trial

I am building an AI-driven situational awareness system that processes multiple live video feeds (RTMP/RTSP) from drones. I need an expert-level engineer to implement the entire NVIDIA Jetson-edge detection pipeline using YOLO + DeepStream, and expose the detection results through an API that can be capture from a client application. I already have a system that has an RTMP container running that is capturing video feeds from the drone so dont worry about that. I just need the YOLO detection pipeline running with the Deepstream API available.

This is a critical component of a larger AI-driven operations product.

If you are a highly skilled DeepStream/Jetson engineer with computer vision and AWS experience, I want to talk to you.

You will take a live RTMP stream coming from a local Docker container on a Jetson device and:
1. Ingest the video via NVIDIA DeepStream SDK
2. Run YOLO detection (people and vehicles) optimized through TensorRT
3. Track objects using DeepStream’s tracking modules (NVDCF)
4. Expose detections through a local API endpoint
5. Bounding boxes
6. Class labels
7. Confidence scores
8. Counts per class (people, cars)
9. Optional zone/ROI metadata
10. Package the entire solution into repeatable Dockerized services on Jetson

Tech Requirements (Must-Have Experience)
You must be strong in the following:

NVIDIA / Jetson Platform
NVIDIA Jetson Nano / Xavier NX / Orin devices
JetPack installation & optimization (L4T, CUDA, cuDNN, TensorRT)
GPU performance tuning on Jetson
Docker on Jetson (ARM64 base images)

NVIDIA DeepStream SDK
Creating DeepStream pipelines for RTSP/RTMP sources
Implementing YOLO as a DeepStream primary-gie (TensorRT engine)
Working with DeepStream probes to extract metadata
Tracking modules (NVDCF / IOU tracker)
Multi-stream handling

Computer Vision / ML
YOLOv8/YOLO11 or similar
Model optimization (FP16, INT8, engine building)
Object tracking
Counting people/vehicles
Understanding of detection noise, false positives, etc.

Backend / API Development
Python or C++ DeepStream bindings
Building a REST/gRPC API to expose detection results
JSON schema design for detection payloads

Cloud (Preferred)
AWS Kinesis, Lambda, API Gateway
Secure pushing of analytics to cloud services
Drone Experience (Preferred but Valuable)
Familiarity with drone video feeds (RTMP/RTSP)

Understanding of MAVLink or telemetry ingestion (not required but helpful)

You will deliver:
1. A working DeepStream pipeline running on my Jetson hardware.
2. YOLO (people + vehicle detection) running via TensorRT.
3. An object tracking layer (DeepStream tracker).
4. An API endpoint to retrieve:
5. Current detections
6. Per-frame counts
7. Optional heatmaps or basic statistics
8. Docker Compose setup for full reproducibility.
9. Documentation on installation, configs, and usage.

Please avoid Chat GPT generated cover letters as I will most likely not select you for interviews if I notice that it was not written by a human.

• Do you have a personal drive to solve complex problems? Have you ever encountered a roadblock with a client, how did you handle it?
• Tell me about yourself on a more personal level (e.g., hobbies, aspirations, and other non-work related things).
• Describe your recent experience with similar projects

We are looking for an experienced mechanical / aerospace engineer to design a custom professional quadcopter platform for industrial use.

The drone will be used for façade cleaning and must replace a DJI Matrice-class drone with a custom, non-DJI platform built from approved components.

This is not a hobby project. The goal is a strong, stable, and professional-looking UAV platform with clean integration of all components.

Key Requirements

* Quadcopter configuration
* 26” carbon fiber propellers
* Very rigid and stable frame (wind-resistant, high thrust)
* Foldable 30 mm arms
* Rear-mounted slide-in battery system
* Target empty weight around ~10 kg
* Payload capability: approx. 5–6 kg
* Professional appearance with integrated housings
(no exposed wiring or DIY-style layout)

Scope of Work

* Overall mechanical layout and architecture
* Integration of all major components
* Foldable arm mechanism
* Payload mounting interface underneath the drone
* Cable routing and mechanical provisions for clean wiring
* Integrated mounts/housings for electronics and sensors
* Structural design suitable for real-world operation
* CAD files suitable for prototyping (CNC / 3D printing / composite parts)

What we are Looking For:

* Experience with multirotor UAVs or aerospace structures
* Strong mechanical integration skills
* Ability to design clean, serviceable layouts
* Understanding of vibration, center of gravity, and structural stiffness
* Experience with carbon/composite structures is a plus

Deliverables

* Complete 3D CAD assembly
* Individual part files
* Prototype-ready design suitable for physical build
* Basic documentation or notes needed to build the prototype

We are looking for a developer or small team to build a minimum viable product (MVP) for 3D volume and dimension measurement using a depth camera in a controlled setup.

This is not a machine learning project.
We are explicitly looking for a deterministic, geometry-based solution.

What we want to achieve
Measure object volume and dimensions from RGB + depth data, with:
• Fixed camera
• Object placed on a flat surface

The goal is a working prototype that proves feasibility and can be extended later.

Technical approach (open to suggestions)
A reasonable approach could involve:
• Depth camera capture (e.g. Intel RealSense or similar)
• Plane detection and object separation using depth data
• Volume estimation via 3D bounding box (minimum requirement)
• Optional: convex hull or mesh-based volume

We are open to alternative stacks as long as the solution is explainable and deterministic.

Output
Preferred output is a simple local REST API (or equivalent) returning:
• Volume
• Dimensions (L × W × H)
• Basic status or confidence indicator

Deliverables
• Working MVP
• Source code
• Basic setup / calibration instructions
• Clear assumptions and limitations

Constraints
• No machine learning
• No cloud dependency required

Hi! I’m the founder of SculptMe App, a consumer-first AI body visualization mobile app. I’m looking for an expert-level AI + mobile developer (or small team) to build an MVP that allows users to upload a full-body photo and receive realistic before-and-after body transformation previews.

MVP scope:
• AI body detection / segmentation
• AI-based body transformation logic (realistic, not filters)
• iOS & Android mobile app with clean, luxury UI
• Backend setup for image processing and storage
• Save and compare results

Required skills:
• AI / Computer Vision / Image Processing
• Experience with body segmentation, pose estimation, or generative models
• Mobile app development (React Native, Flutter, Swift, or Kotlin)
• Backend & cloud infrastructure (AWS, Firebase, etc.)

How I like to work:
• Fixed-price project with clear milestones
• Start with a paid discovery phase
• Clear communication and regular updates
• Open to technical recommendations

This is an early-stage startup with long-term potential. Looking forward to thoughtful proposals.

— Zakiyyah
Founder, SculptMe App