Комп'ютерний зір Jobs

I want to build a small, self-contained charging station that wakes up only when it “sees” a bottle and a can. The core will be an Arduino (any recent 32-bit board is fine) tied to a solar panel and battery pack. A lightweight AI vision module or camera running something like TensorFlow Lite or OpenCV should continuously—or on a duty-cycled basis—look for bottles. The moment a bottle is detected, the system must move straight into the charging routine without waiting for external confirmation; no alerts, no logging, just automatic activation. What I need from you • Complete Arduino firmware that integrates the vision trigger with the charge-controller logic. • A schematic / wiring diagram showing how the camera, solar panel, charge controller, and any...

I have continuous video footage from live inspection of steel flat bars strips our production line and need a complete deep-learning pipeline that flags surface, structural, and functional defects in real time. The raw videos are already labeled by timestamp; frame-level annotation may still be required for optimal accuracy. Scope • Design and train a deep neural network—CNN, transformer, or hybrid model—that detects all three defect categories directly from video streams. • Implement preprocessing (frame extraction, augmentation, ROI isolation) and post-processing (tracking, alert generation) in Python using libraries such as PyTorch/TensorFlow and OpenCV. • Optimise for inference on an on-premise GPU; latency under 200 ms per frame is the target. &bull...

I need a single AI application that can see, hear and speak to the user. Using my own OpenAI key (or, if you prefer, a Gemini or Claude endpoint), I want you to wire conversational logic with the device camera so the assistant can recognise whatever the lens captures—faces, emotions, objects, actions, text, you name it—then hold a natural dialogue about what it sees. The build has to run everywhere: a mobile version for iOS & Android, a web app that works in the browser, and a desktop release for Windows and macOS. Users should be able to create an account, log in, and start interacting immediately. Speech-to-text converts their voice to prompts, vision models process the live feed, and text-to-speech delivers the reply in real time. For LLM calls, default to ChatGPT via ...

I’m building a ROS 2 pipeline that must reliably spot nuts and bolts in a live camera feed and publish their positions to the rest of my stack in real time. Your job is to create the complete vision-detection module—from model training or fine-tuning through to a clean ROS 2 node that subscribes to an image topic and spits out the detected objects with bounding boxes (or masks) and a confidence score. OpenCV, TensorFlow/PyTorch and any of the common ROS 2 image-transport plugins are all fine as long as the final node runs on Humble and stays GPU-agnostic (CUDA acceleration is a bonus, not a requirement). I already have a test rig with a standard USB camera; if you need specific calibration images I can capture them for you. Please deliver: • Source code for the detection...

I would like help finishing a program I started using Cursor AI. I want to be able to use a 3d stl file of the bottom of foot and overlay an image of a foot orthotic to match the bottom of the foot.

我正在为一款工业相机进行 ISP 图像测试与色彩调试，目标是在 1 个月内交付可量产的调试参数。任务仅针对静态图像，核心是通过反复测试和优化，让相机在多种光照与色温条件下都能还原准确、自然的色彩。 主要工作内容 • 制定静态图像测试方案：包括分辨率卡、人像卡、灰阶卡等常规靶标及光源条件 • 采集 RAW 与 YUV / JPEG 对比样张，输出测试报告 • 根据测试结果微调白平衡、色彩矩阵、Gamma 等 ISP 模块参数 • 给出最终可直接烧录的参数表，并附带调试说明书 完成标准 1. 色彩 ΔE 满足既定工业相机参考值 2. 典型场景（室内、室外、弱光）均通过对比评审 3. 提交完整测试数据、调试过程记录及最终参数 时间要求 全部工作须在 1 个月内完成，并支持 1 次远程复测协助。 我更看重相关经验。请在申请中突出您以往做过的相机 ISP、色彩或图像质量调优案例，以及常用工具链（如 IQ Tuning, Imatest, X-Rite 或自研脚本等）。如果能在同行业项目中展示过可衡量的提升效果，将优先考虑。

I'm looking for a proficient image capturing and processing system to ensure quality control of our products. The primary focus is on maintaining color consistency. Key Requirements: - Capture high-quality product images for quality control. - Process images to detect and signal color consistency issues. Ideal Skills: - Experience in image processing and computer vision. - Familiarity with quality control processes in manufacturing. - Ability to develop signaling mechanisms for quality deviations.

I need a small software tool that runs on an NVIDIA Jetson Orin Nano to capture, organize, and export a labeled image/video dataset for training YOLO object detection models. The camera is an 640×512 sensor connected via MIPI CSI-2 (4 lanes) using a 22-pin ribbon cable to the Jetson Orin Nano CSI connector. The tool should support live preview, capture, dataset management, and YOLO-format export. Build an application that: - Interfaces with the camera on Jetson Orin Nano - Acquire frames from the MIPI CSI-2 camera reliably (GStreamer / V4L2 / libargus or best approach for this sensor). - Provide live preview (grayscale or false color optional). - Display FPS + resolution. Dataset capture: - Capture single frames and/or short clips (optional). - Save images in a consist...

More details: Which deep learning framework do you prefer for this project? TensorFlow Do you have a preferred dataset for brain MRI image segmentation? Please use a publicly available dataset Which style of output visualization do you prefer? 2D slices with segmentation overlay

I need a detailed physics-based simulation for autonomous robots operating in urban environments. The simulation should incorporate the following sensors: LIDAR, Camera, and Ultrasonic. Key Requirements: - Development of a realistic urban physics-based simulation. - Integration of LIDAR, Camera, and Ultrasonic sensors. - Focus on urban environment dynamics and challenges. Ideal Skills and Experience: - Expertise in simulation development, particularly physics-based. - Experience with sensor integration - Strong background in robotics and urban environment modeling. - Proficiency in relevant simulation software and programming languages.

I need a small utility that hooks into a PC game built on a custom engine and, at run-time, captures three things per frame: • the full 4×4 view-projection camera matrix • the depth buffer in linear space • the raw, uncompressed RGB back-buffer All data should be written straight to disk in a clean, well-documented CSV schema so I can feed it directly to my AI training pipeline later. Resolution, frame interval and output folder must be user-configurable through simple flags or an .ini file. I am comfortable providing you with a private build of the game plus symbols; the rest of the reverse-engineering or graphics API interception logic is yours to implement. Feel free to rely on tools such as RenderDoc, NVIDIA Nsight, Detours, or your own DLL injection met...

Real-Time Face Swap System Requirements Specification 1. Project Overview Purpose: Development of real-time face replacement functionality aimed at protecting privacy and expanding expressive capabilities during streaming. Users: Exclusively for individual streamers. Streaming Environment: - Current: Mac mini (M4 chip, 16GB RAM) - Future: Upgradeable to 32GB/64GB as needed - OS: Mac preferred (portability priority), Windows compatibility considered if necessary - Streaming Software: OBS to be used 2. Non-Negotiable Requirements (Must Have) 2-1. Technical Requirements Face Replacement Specifications: - Replace only eyes and eyebrows - Use original parts for nose down (mouth/lips) - Reason: To prevent double lips when drinking liquids Safety: - Face processing must never drop out w...

I’m building a diffusion–based pipeline focused on image generation, and I want the results to be crisp, large-format visuals rather than the low-res outputs many models settle for. Your task is two-fold: first, curate or assemble a fit-for-purpose dataset (I don’t yet have one), and second, train a state-of-the-art z-image diffusion model that consistently produces high-resolution renders. You’ll have freedom in tool choice—PyTorch, TensorFlow, DreamBooth, LoRA or any other modern techniques are fine—as long as the final model can be reproduced from the training scripts you deliver. I expect the usual artefacts: cleaned dataset (with clear licensing notes), training code, model checkpoints, and a concise README outlining hyper-parameters, compute used,...

I’m putting together a straightforward online page where visitors can upload a photo and immediately receive automated face-based feedback powered by AI image recognition. The core requirement is reliable detection of faces in any uploaded image; once a face is located, I’d like the response to display bounding boxes or similar visual confirmation so users clearly see what the model found. Beyond that essential functionality, I’m open to expanding into deeper facial insights—age, emotion, gender, or related analytics—if you have an off-the-shelf model or a custom pipeline that can slot in cleanly. My priority, however, is to ship a polished, latency-friendly MVP that runs either server-side (Python + OpenCV, TensorFlow, or similar) or directly in the browse...

I am building a fully-autonomous vacuuming robot around a Jetson Nano and a 3-D LiDAR and need an experienced ROS 2 developer to take it from a proof-of-concept chassis to a reliable household machine. Core capabilities I’m after • Vacuuming as the single cleaning function • ROS 2 Navigation stack for path planning, live re-planning and return-to-dock • Real-time obstacle avoidance fused from the LiDAR and the onboard IMU • SLAM or comparable mapping so the robot can save, edit and reload floor maps between sessions The hardware is already on hand: Jetson Nano, 3-D LiDAR, motor controller, wheel encoders, IMU and a basic suction module. Your job is to wire up the software side, tune the parameters for smooth coverage, and expose a simple YAML/JSON config...

I’m putting together a straightforward online page where visitors can upload a photo and immediately receive automated face-based feedback powered by AI image recognition. The core requirement is reliable detection of faces in any uploaded image; once a face is located, I’d like the response to display bounding boxes or similar visual confirmation so users clearly see what the model found. Beyond that essential functionality, I’m open to expanding into deeper facial insights—age, emotion, gender, or related analytics—if you have an off-the-shelf model or a custom pipeline that can slot in cleanly. My priority, however, is to ship a polished, latency-friendly MVP that runs either server-side (Python + OpenCV, TensorFlow, or similar) or directly in the browse...

I need a compact, fast object-detection model that runs directly on an edge board (Jetson Nano, Raspberry Pi, Coral or similar) and processes aerial images from drones. The immediate application is surveillance, yet the solution should stay flexible enough to be reused later in agriculture or disaster-response scenarios. Source imagery will contain a mix of people, cars, buses, bicycles and assorted infrastructure. The model must be especially reliable at spotting people and critical infrastructure elements, while still recognising the wider vehicle classes.I am open to any justified architecture (YOLOv8, MobileNet-SSD, EfficientDet-Lite, or superior alternatives)—provided it outperforms YOLOv9c and similar models while delivering real-time inference on edge devices once quantized o...

Project Overview We are developing an advanced AI-powered camera monitoring system for hotel reception areas to prevent revenue leakage, track staff activity, and create intelligent customer profiles. The system will integrate with CCTV cameras and use AI-based face recognition to monitor visitor interactions, identify revenue mismatches, and generate actionable reports. We are looking for an experienced AI developer, computer vision expert, or full-stack AI team who can build and deploy a production-ready solution. ⸻ Main Objectives • Detect every visitor entering reception • Recognize hotel staff members • Automatically create customer profiles • Track complete customer–staff interaction history • Prevent revenue theft and unrecorded bookings &bu...

I need a deep-learning solution that watches a driver’s face through a standard camera feed, tracks eye-closure patterns and yawning frequency, then translates those cues into a clear fatigue score that updates continuously. Over a journey the model should also plot a time-based curve so I can see how alertness rises or falls. Please build and train the full pipeline in Python, preferably with PyTorch or TensorFlow paired with OpenCV for video handling. The system must be completely vision-based; no wearables or contact sensors. I will supply sample clips for initial testing, but the code should accept any 30 fps video stream so I can later attach it to an in-car webcam. The final hand-off should include: • Inference script that ingests a live or recorded feed, detects eyes ...

I need an end-to-end system that automatically counts every passenger who enters or exits a bus and, at the same time, lets only authorised riders board through face recognition. The headcount must update live and stream to my security team’s server so they can monitor occupancy and verify that no unauthorised person is on board at any moment. Here is what matters to me: the facial verification must be fast enough to avoid boarding queues, the counting accuracy should stay above 98 %, and the data flow has to be truly real-time, not batch. If you already work with edge cameras or depth-sensing devices for people counting, please tell me how you will integrate them with your face-recognition engine and how you plan to push the feed straight to our internal server (REST API, MQTT, or ...

I need a deep-learning solution that watches a driver’s face through a standard camera feed, tracks eye-closure patterns and yawning frequency, then translates those cues into a clear fatigue score that updates continuously. Over a journey the model should also plot a time-based curve so I can see how alertness rises or falls. Please build and train the full pipeline in Python, preferably with PyTorch or TensorFlow paired with OpenCV for video handling. The system must be completely vision-based; no wearables or contact sensors. I will supply sample clips for initial testing, but the code should accept any 30 fps video stream so I can later attach it to an in-car webcam. The final hand-off should include: • Inference script that ingests a live or recorded feed, detects eyes ...

I want to build a production-ready image-recognition solution and I need an AI specialist to make it happen. The broad goal is clear—accurate, real-time image recognition—but I’m still weighing which specific task (object, facial, or text detection) will bring the most value to my workflow. Your first job will be to help me evaluate these use-cases and settle on the one that fits my data, timeline, and performance targets. Once we lock the scope, I’ll provide sample images plus any domain knowledge you need. You’ll handle the end-to-end pipeline: dataset preparation, model architecture selection (CNN, transformer, or whatever you judge best), training/tuning, and deployment in a lightweight API or on-device model. I’m comfortable with common stacks like...

I want to build a production-ready image-recognition solution and I need an AI specialist to make it happen. The broad goal is clear—accurate, real-time image recognition—but I’m still weighing which specific task (object, facial, or text detection) will bring the most value to my workflow. Your first job will be to help me evaluate these use-cases and settle on the one that fits my data, timeline, and performance targets. Once we lock the scope, I’ll provide sample images plus any domain knowledge you need. You’ll handle the end-to-end pipeline: dataset preparation, model architecture selection (CNN, transformer, or whatever you judge best), training/tuning, and deployment in a lightweight API or on-device model. I’m comfortable with common stacks like...

I want to learn how to do annotations using CVAT. I have a project in my mind and it requires CVAT for annotating so you will be required to guide me as per my project requirements. I need someone to start me with the basics and i will pick up form there. classes will be hourly and recorded.

I want to build a production-ready image-recognition solution and I need an AI specialist to make it happen. The broad goal is clear—accurate, real-time image recognition—but I’m still weighing which specific task (object, facial, or text detection) will bring the most value to my workflow. Your first job will be to help me evaluate these use-cases and settle on the one that fits my data, timeline, and performance targets. Once we lock the scope, I’ll provide sample images plus any domain knowledge you need. You’ll handle the end-to-end pipeline: dataset preparation, model architecture selection (CNN, transformer, or whatever you judge best), training/tuning, and deployment in a lightweight API or on-device model. I’m comfortable with common stacks like...

I want to build a production-ready image-recognition solution and I need an AI specialist to make it happen. The broad goal is clear—accurate, real-time image recognition—but I’m still weighing which specific task (object, facial, or text detection) will bring the most value to my workflow. Your first job will be to help me evaluate these use-cases and settle on the one that fits my data, timeline, and performance targets. Once we lock the scope, I’ll provide sample images plus any domain knowledge you need. You’ll handle the end-to-end pipeline: dataset preparation, model architecture selection (CNN, transformer, or whatever you judge best), training/tuning, and deployment in a lightweight API or on-device model. I’m comfortable with common stacks like...

I need a working proof-of-concept that spots drones in real time by blending vision and RF cues. On the vision side you will train and fine-tune YOLOv8 with Anti-UAV and VisDrone footage. In parallel, spectrograms coming from DeepSig RadioML together with my own SDR captures (HackRF / RTL-SDR) should feed a CNN that flags drone-class emissions. Once both streams run reliably, I want them merged—either with a straightforward rule set or a small neural fusion layer; I’m happy to discuss which choice achieves the best balance of speed and robustness. The finished model must be benchmarked on four fronts that matter equally to me: Accuracy, Precision, Recall, and False-Alarm Rate. Because the project will eventually guard a sensitive site, I’m aiming for a solution that can...

Будь ласка, зареєструйтесь або увійдіть в систему для перегляду деталей.

I'm looking for a pure Python RPA expert with a strong background in NLP and computer vision libraries/ models. Your role will involve web scraping, designing neural networks, and automating complex workflows. Patience and a proven track record in alpha testing are essential. Key Requirements: - Proficiency in Python with RPA tools - Experience with web scraping using playwright or Selenium - Designing and implementing neural networks - Expertise in Vision libraries/offline models (without using API), NLP using spaCy - Proficient in computer vision with TensorFlow. Preferred Skills: - ML Ops and training knowledge - Strong problem-solving skills - Ability to handle complex workflows Ideal Candidate: - Proven experience and portfolio - Patient and methodical approach Looking forwar...