Développement d'Application Vision intelligent

@MashoodurRehman1

I am a highly skilled software developer with extensive experience in computer vision, specializing in Python and C++. I possess a strong track record of building real-time visual analysis systems, making me well-suited for the 'Développement d'Application Vision intelligent' project. My proficiency in Python for interface development, combined with deep expertise in integrating high-performance C++ modules via pybind11, aligns perfectly with the project's technological architecture. My experience includes developing multi-source detection systems using OpenCV and creating efficient object tracking solutions that maintain a persistent identification across multiple camera inputs. I am adept at implementing spatial estimation techniques and have prior experience with facial recognition using LBPH, ensuring accurate re-identification in various scenarios. Additionally, I have created customizable user interfaces that enhance the functionality and ease of use, similar to your project's requirements. I am keen to explore how my skills and experience can contribute to the success of your application. Could we schedule a discussion to delve deeper into your specific needs and goals?

@Mahihassi

As a talented Electrical Engineer, I have proven expertise in developing robust firmware and software solutions complete with intelligent connectivity - a skillset that aligns itself perfectly with your vision for the Resistance High Vision application. Having mastered Python and C++, I am skilled in leveraging the strengths of both languages similar to your project's architecture. My extensive experience with STM32, ESP32, Rasberry Pi, and Arduino will further bolster the development and convergence of your next-gen vision tracking system. Additionally, my aptitude in PCB design, RF Hardware development, AI Integration (including Machine Learning and Deep Learning) will drive crucial elements like 3D modeling for enclosures, data analytics, and even mobile/web app functionality responsible for managing and monitoring video feeds. Building on my background integrating PLCs and SCADAs, I am also well-suited to integrate this application into larger industry control systems. Lastly, my interdisciplinary approach to product development - from concept to final product - matches your full-service requirements. Derived from dedicated work as a freelancer over 6 years, my adaptable problem-solving nature has led me to develop applications applicable in fields like Biomedicine.

@corpmember29

Bonjour, concernant votre application de vision intelligente : {{{ J'ai déjà développé des applications de bureau similaires basées sur la vision par ordinateur et l'IA, et je peux vous en faire la démonstration. }}} J'ai étudié attentivement votre cahier des charges et je comprends la nécessité d'une application de vision en temps réel prenant en charge la détection multi-caméra, le suivi persistant des objets, l'estimation de distance, la reconnaissance faciale, l'enregistrement ainsi que l'historique local des détections. Je possède plus de 13 ans d'expérience dans les technologies requises ; je maîtrise Python, OpenCV, C++, pybind11, Tkinter, les modèles de vision par ordinateur, le suivi d'objets, le stockage local de données et le packaging d'applications de bureau pour Windows, macOS et Linux. Je suis en mesure d'implémenter l'interface multi-caméra, la détection d'objets (68 catégories) avec basculement sur détection de mouvement, les identifiants persistants, l'estimation spatiale, la reconnaissance faciale, l'enregistrement avec annotations, les journaux au format JSON, l'exportation en GIF, la persistance des données locales ainsi que le mécanisme de secours en Python si le module C++ n'est pas disponible. JE FOURNIRAI UN SUPPORT CONTINU GRATUIT PENDANT 2 ANS AINSI QUE LE CODE SOURCE COMPLET. Je peux également fournir des scripts de compilation documentés et une documentation d'installation pour les plateformes de bureau concernées. Cordialement, Christina

@maxscender

Votre besoin est un vrai système temps réel, pas une maquette : multi-caméras, noyau C++ critique, persistance des visages et packaging multi-OS. Je sécurise d’abord la brique pybind11 avec un wrapper C++ compact et testable, tout en gardant un fallback Python équivalent pour que rien ne bloque si le build échoue. Je structure le pipeline vidéo avec un thread dédié par webcam, un tracker à identifiants persistants, et une base LBPH 100% locale. Je gère aussi les pièges concrets : noms réels des caméras USB, concurrence avec Tkinter, coût des calculs géométriques, et compilation finalisée pour Windows, macOS et Linux. Je livre un code propre, documenté, avec le guide d’installation et la voie Android via Chaquopy.

@PINESCRIPTERR

Le principal point de performance ne sera probablement pas le calcul géométrique en C++, mais le décodage simultané de six flux et l’inférence du détecteur. Je structurerais donc le pipeline avec des files bornées par caméra, un planificateur d’inférence partagé et l’abandon contrôlé des images anciennes afin de conserver une interface fluide sans accumuler de latence. Je peux stabiliser l’architecture Python/C++ autour de modules clairement séparés : acquisition, détection, suivi, visages, estimation spatiale, stockage et interface. Le suivi persistant devra distinguer identifiant de trajectoire et identité faciale, gérer les disparitions temporaires, puis éviter les doubles comptages. Les écritures vidéo et JSON fonctionneront dans des threads dédiés, avec rotation des journaux et chiffrement local optionnel pour les données faciales. Je prévoirais également des tests avec plusieurs caméras, le repli Python sans extension C++, des builds reproductibles Windows/macOS/Linux et des mesures réelles de FPS, latence, mémoire et stabilité. LBPH resterait présenté comme une ré-identification locale, avec consentement et suppression des données. Quel matériel minimum et quel débit cible souhaitez-vous garantir avec six caméras actives simultanément ? Cordialement, Houssame

@mayr81

Développement d’une application vision intelligente bureau en Python (Tkinter) avec cœur calcul C++ via pybind11 : détection multi-source, suivi persistant, estimation spatiale monoculaire et reconnaissance faciale locale (LBPH) sans cloud. Implémentation d’un pipeline OpenCV pour la détection 68 catégories, gestion multi-aperçu jusqu’à 6 flux, et traçage stable avec identifiants uniques. Ajout d’un module distance_core en C++ pour accélérer les calculs répétitifs, tout en garantissant un repli Python si compilation indisponible. Intégration de la persistance entre sessions (visages mémorisés), journaux JSON continus indépendants du bouton d’enregistrement, export GIF animé à la demande et capture vidéo annotée. Livraison : code source complet, documentation, scripts de compilation pour Kali Linux (.AppImage), macOS (.dmg) et Windows (.exe), guide d’installation incluant portage Android via Chaquopy. Format et architecture prêts pour itérations rapides : interface et orchestration côté Python, performance native côté noyau C++.

@toriquldev123

Hello There! I’m Md Toriqul Islam, an experienced Python/C++ developer specializing in computer vision, real-time video processing, desktop applications, and performance-focused architectures. I’m excited to work on Resistance High Vision. I have rich experience in Python, C++, OpenCV, pybind11, computer vision, object detection, tracking, facial recognition, multi-camera processing, JSON logging, and cross-platform application development. I am skilled in Python, C++, OpenCV, Tkinter, pybind11, object tracking, facial recognition, video processing, native performance optimization, and Linux/Windows/macOS deployment. I understand you need a professional real-time computer vision application supporting up to six webcams, persistent object tracking, distance/speed estimation, face recognition, local data persistence, recording, JSON logs, and a Python/C++ hybrid architecture with a reliable Python fallback. I’m ready to start immediately and can work carefully within the stated limitations, keeping all processing local and maintaining clean, documented, cross-platform code. Looking forward to hearing from you. Best regards, Md Toriqul Islam

@mahig47

Bonjour ! Comme indiqué dans votre annonce, vous souhaitez développer Resistance High Vision, une application de bureau professionnelle de vision par ordinateur en temps réel capable de transformer plusieurs webcams en un système d'analyse visuelle intelligent. Le projet combine la détection d'objets, le suivi persistant, l'estimation spatiale, la reconnaissance faciale, l'enregistrement et l'historique de détection continu au sein d'une interface de bureau personnalisable. Je me concentrerai sur la création de l'interface applicative et de la couche d'orchestration en Python, tout en intégrant le noyau de calcul C++ via pybind11 pour les opérations numériques critiques en termes de performances. Je peux implémenter la découverte et la prévisualisation multicaméra, la détection d'objets dans les catégories requises, les identifiants de suivi persistants, l'estimation de distance et de mouvement, la reconnaissance faciale avec profils persistants. Je suis spécialisé en Python, C++, vision par ordinateur, OpenCV, détection et suivi basés sur l'IA, développement d'applications de bureau, intégration de caméras et optimisation des performances. Contactons-nous pour examiner l'architecture technique existante, les modèles, les exigences relatives aux caméras et les composants de calcul C++ afin de construire un système de vision stable et performant autour de votre concept actuel. Cordialement, Nikita Gupta

@vladang

Six flux simultanés sans bloquer Tkinter seront le point critique de Resistance High Vision. Le suivi persistant, la mémoire locale et le repli Python/C++ doivent aussi rester cohérents après une perte de caméra ou un redémarrage. Je commencerai par valider la capture multi-source et le cycle complet détection → suivi → journal JSON. Chaque caméra aura une file d’images limitée pour éviter l’accumulation et garder l’interface fluide. Le noyau pybind11 libérera le GIL pendant les calculs C++, avec des tests comparant ses résultats au repli Python. Je vérifierai aussi les écritures atomiques des visages et journaux pour éviter une base corrompue après un arrêt brutal. Le code existe-t-il déjà pour audit et finalisation, ou faut-il construire l’application complète depuis cette architecture ?

@demivision

Bonjour, Votre projet Resistance High Vision correspond parfaitement à mon expérience en vision par ordinateur et applications desktop Python/C++. Je peux travailler sur l’architecture existante, notamment OpenCV, détection et tracking multi-caméras, estimation de distance, reconnaissance faciale, traitement vidéo et intégration C++/pybind11. Je suis également à l’aise avec l’optimisation des performances temps réel, la gestion de plusieurs flux webcam, la persistance des données locales et la création d’applications distribuables sous Windows, macOS et Linux. Je comprends également l’importance de conserver un fallback Python lorsque le module C++ n’est pas compilé, afin de garantir la fiabilité de l’application. Je serais heureux d’examiner votre code source, d’identifier les points à améliorer et de contribuer aux prochaines étapes du développement. Cordialement.

@PrateekP3003

40 hours/week, available for work. You can track project progress via the tracker. Hi There! I'm a full-time software developer with 7+ years of experience in Python, C++, computer vision, desktop applications, and AI-driven systems. Your Resistance High Vision architecture is particularly aligned with my experience, especially the Python/OpenCV orchestration with a performance-critical C++ layer through pybind11. I can work on: Multi-camera webcam discovery and up to 6 simultaneous video streams OpenCV-based object detection, tracking, and fallback motion detection Persistent object IDs and optimized tracking pipelines Distance, dimensions and approach/recede speed calculations Face detection/recognition with local persistent storage JSON detection logs, annotated video and GIF export Tkinter UI, settings, themes and real-time performance optimization C++/pybind11 integration with a reliable Python fallback Windows .exe, macOS .dmg and Linux/AppImage packaging Clean local-only architecture with no unnecessary cloud/network dependency I can also review the existing codebase rather than rebuilding working components, identify CPU/memory bottlenecks, improve multi-camera stability and ensure the C++ module fails gracefully when unavailable. If you can share the current source repository and target platforms, I can review the architecture first and propose the most practical implementation and optimization plan. Looking forward to collaborating with you!

@mughiraaa

Hello, I have already completed similar computer vision and Python applications involving OpenCV, object detection, tracking, camera streams, and local data processing. I have strong experience in web development, APIs, and scalable software solutions, so I can quickly understand your Python and C++ architecture and deliver exactly what is needed. I can work on the multi camera interface, persistent tracking, distance estimation, face recognition, local storage, video recording, JSON logs, and the pybind11 C++ integration while keeping the Python fallback available. Will the 68 object categories and existing detection models already be provided, or should I also handle model selection and integration? I would be happy to review the current code, discuss the architecture, and schedule a quick meeting to get started. I will share my portfolio in chat I look forward to hear from you. Thanks Best Regards, Mughira

@dataspro

Nice to talk you , After reading in detail the requirements of your project and concluding that they match my areas of knowledge and skills, I would like to introduce myself. My name is Anthony Muñoz and I am the lead engineer for DS Pro IT agency. I have worked for over 10 years in Backend and software development and have successfully done multiple jobs. It will be a pleasure to work together to make your project a reality. Please feel free to contact me. I´m looking forward to working with you. I really appreciate your time and remain attentive to any request or question. Greetings

@shayankhan333

Hi, I can build Resistance High Vision as a Python and C++ desktop app with pybind11, covering multi-webcam detection, persistent tracking, distance estimation, and LBPH facial recognition, all fully local. I can start today. I will also add the Python fallback for distance_core so the app runs even when the C++ module is not compiled. Questions: 1) Which OS build is the priority first: Windows, macOS, or Kali AppImage? 2) Should the 68-object model ship preloaded, or configurable by the user? Looking forward to discussing further. Regards, Shayan.

@invoketech

This is a real-time computer-vision desktop application that turns multiple webcams into a local visual-analysis system. The main development work is Python/OpenCV for the UI and orchestration, with C++/pybind11 for performance-critical spatial calculations. Core Features: Multi-camera detection with up to 6 webcams 68-category object detection Persistent object tracking & counting Distance, size and movement-speed estimation Local face recognition & face database Annotated video recording & GIF export Continuous JSON detection/history logging Python fallback when C++ module is unavailable Tkinter desktop interface with customizable themes 100% local processing with no cloud dependency Windows, macOS & Linux packaging Complete Python + C++ source code and documentation Architecture: Webcams → OpenCV Detection → Object Tracking / Face Recognition → C++ Spatial Core → Tkinter UI → Local Database & Logs This is best approached as a desktop computer-vision engineering project, with careful optimization around multi-camera processing, persistent tracking, and reliable cross-platform packaging.

@juanestebana17

Hello! @Project@ I read your project and understand that you need a professional computer vision application with real-time object detection, face recognition, and spatial estimation, leveraging Python and C++. @Why I'm a good fit@ I have extensive experience in software development with Python, OpenCV, and object detection, along with expertise in C++ for performance-critical components. For years, I’ve developed similar vision-based applications focusing on real-time analysis and face recognition, ensuring both accuracy and efficiency. @A quick question:@ Are there specific hardware constraints or camera specifications we need to consider? I'm ready to start working immediately. Thanks!

@SkillCrafts

I understand you're building a sophisticated real-time computer vision application, "Resistance High Vision," leveraging Python for UI/orchestration and C++ for performance-critical components via pybind11. My experience developing similar desktop applications with integrated C++ backends for demanding computational tasks, such as optimizing image processing pipelines and implementing real-time object tracking algorithms, directly aligns with your project's core requirements. My technical approach would involve building upon your existing architecture. I would focus on optimizing the C++ core for specific vision tasks like object detection and facial recognition, ensuring efficient data flow and minimal latency. This would include utilizing libraries like OpenCV for image manipulation and potentially integrating deep learning frameworks (e.g., TensorFlow Lite or PyTorch Mobile if suitable for desktop deployment) for enhanced accuracy. I would also ensure seamless integration with your Python interface for a fluid user experience. To ensure we're perfectly aligned, could you elaborate on the specific performance metrics you're targeting for each vision module (e.g., FPS for object detection, accuracy for facial recognition)? Additionally, what is your envisioned deployment environment for this desktop application? I'm eager to discuss how I can contribute to making Resistance High Vision a robust and high-performing solution.

@jitendras097

Hello, I can build Resistance High Vision as a professional real-time computer vision desktop application using Python + OpenCV with a C++/pybind11 performance layer. I’ll implement multi-camera detection, persistent object tracking, distance/speed estimation, local face recognition, recording, JSON logging and the customizable desktop interface while keeping all data completely local. Key Features: Multi-Webcam Support | 6-Camera Preview | Object Detection | Persistent Tracking | Distance/Speed Estimation | Face Recognition | Video/GIF Export | JSON Logs | Local Database | Tkinter UI | C++/pybind11 | Windows/macOS/Linux. I’ll keep the C++ module optional with a Python fallback, provide clean documented source code, build scripts and installation documentation for deployment on the required platforms. Thanks, Jitendr

@rezamahi

I’d be very interested in developing and improving Resistance High Vision. Your architecture is especially interesting because it combines Python for UI/orchestration with a performance-focused C++ core through pybind11. I can work across the full pipeline: OpenCV-based multi-camera capture, object detection, persistent multi-object tracking, face recognition/storage, spatial estimation, speed calculation, JSON logging, annotated video/GIF export, and the Tkinter interface. I can also ensure the C++ module has a clean Python fallback so the application remains functional when the native extension is unavailable. I’ll keep the code modular and maintainable, with clear separation between the UI, detection/tracking, face database, logging, and C++ computational layer. I’m comfortable working with Python, C++, OpenCV, NumPy, pybind11, computer vision pipelines, and desktop application development. I can also help with testing across multiple webcams and preparing the required Windows/Linux/macOS builds. I understand the limitations you specified around monocular distance estimation and LBPH recognition and will keep the implementation technically honest rather than overclaiming accuracy. I’d be happy to review the existing source and start with the highest-priority components. Best regards, Reza