Find the best FFmpeg jobs

I am looking for help with a computer vision project. It is mostly available on a GitHub page, but I need some modifications and additions to meet my needs. The application is based on "Multiply" from GitHub. It currently works with sequences involving 2+ actors, but I need it to work with sequences involving a single actor as well as multiple actors. I would prefer the application to work with JPG sequences only. FFmpeg integration is not needed. The preprocessing stage must include the ability to export camera and SMPL-X data to FBX format and also to import camera and SMPL-X data from FBX. This is a must-have, as opposed to being limited to forced trace input → training. We can use the example data available on GitHub to start, and I can also provide data for single-actor sequences. For preprocessing, I want to integrate ViTPose and OpenPose to improve accuracy. FBX import and export should be the preferred method for handling camera and SMPL-X data. I will handle the custom SMPL-X Maya rig and cameras on my side. If the trace camera and SMPL-X data can be exported to me as FBX (from the preprocessing stage), I will set up my system to parse the data into the trainer. The goal is to ensure compatibility with my workflow, allowing seamless export/import of camera and SMPL-X data via FBX while enhancing the accuracy of preprocessing with ViTPose and OpenPose.

NO AGENCIES PLEASE I am looking for developers (more than one) for several ongoing projects, requiring skills in writing algorithms for Looking for a Python ML/CV developer who has experience in the following areas: - OpenCV - camera calibration and 3D reconstruction - streaming video processing (GStreamer, ffmpeg experience would be a plus) - Tensorflow and Keras (convolutional neural networks) - examples of your code in any of the areas would be a huge advantage It would be an advantage: - Published scientific articles related to the ML/CV field - Python multiprocessing/multithreading advanced techniques - C++/Python integration - PyTorch - Camera calibration - Objects detection - Objects 3D position estimation - Streaming video (GStreamer, ffmpeg) I am looking for someone who is communicative, gives suggestions, asks questions, and understands the product delivery requirements. While responding, please write about the kind of AI projects you have done with the features and some insights on your experience with Camera calibration, Objects detection, object 3D position estimation, Streaming video (GStreamer, ffmpeg)

We are looking for a skilled developer to create a command-line interface (CLI) tool that adds video stabilization as a post-production step. The tool should utilize FFMPEG (or something else) to stabilize shaky videos effectively. The ideal candidate will have experience with video processing and familiarity with FFMPEG commands and video formats. This project is crucial for enhancing video quality in our production workflow. If you have a passion for video editing and post-production tools, we would love to hear from you! Our cameras can be attached to a wall, pole, crane... There may be cameras 100% static, others affected only by vibrations and others will be swinging in the space if they are attached to a crane. The algorithm to stabilize must take care of this or use a parameter that we can define for each job.

We are seeking an experienced developer to build a custom WebRTC video chat server capable of recording video streams using Node.js on Ubuntu Linux. Will consider use of MediaSoup library. The ideal candidate will have a solid understanding of WebRTC, video streaming and server architecture. You will be responsible for designing and implementing a reliable system that ensures high-quality video communication and effective recording features. If you have a strong background in real-time communication technologies, we'd love to hear from you!

Zoom AI Eye-Contact Correcting App 1. Project Overview • Objective: Develop an AI-powered app for Zoom that corrects eye contact by adjusting the user’s video stream in real-time. • Platform: Desktop application or Zoom-integrated app via the Zoom App Marketplace. • Core Features: 1. Real-time face and eye tracking. 2. AI-based video adjustment for natural eye contact. 3. Seamless integration with Zoom or functionality as a virtual camera. 4. User-friendly interface with toggles and calibration settings. ________________________________________ 2. Technology Stack • Frontend: o Framework: React, Electron.js (for desktop app). o Tools: HTML, CSS, JavaScript. • Backend: o Language: Python (for AI processing), Node.js (for server logic). o Frameworks: Flask/Django (Python) or Express.js (Node.js). • AI/Computer Vision: o Libraries: OpenCV, MediaPipe, Dlib. o AI Models: Pre-trained models for face and eye tracking (e.g., AffectNet, GazeCapture). • Real-Time Video Processing: o Technology: WebRTC for video streaming. o Tools: FFmpeg for video manipulation (if required). • Zoom Integration: o SDKs: Zoom Video SDK, Zoom Meeting SDK. o API: Zoom’s REST APIs for app integration and user authentication. • Database: o Firebase or MongoDB for storing user preferences (optional). • DevOps: o CI/CD Tools: GitHub Actions, Jenkins. o Cloud Hosting: AWS, Azure, or Google Cloud for backend services. ________________________________________ 3. Core App Features 1. Real-Time Eye Contact Adjustment: o Detect user’s face and eye position using computer vision. o Adjust video feed to simulate direct eye contact with AI-driven transformations. 2. Seamless Zoom Integration: o Use the Zoom Video SDK to access and manipulate video streams. o Offer options to enable or disable eye-contact correction during meetings. 3. User Interface: o Controls to toggle eye-contact correction. o Calibration options for individual preferences. o Live preview of video feed with corrections applied. 4. Low-Latency Video Processing: o Optimize video processing to ensure minimal lag during live meetings. 5. Privacy Protection: o Local processing of video streams to ensure user privacy. o No storage of user video or data unless explicitly authorized. ________________________________________ 4. Development Timeline • Phase 1: Research & Prototyping (2 weeks) o Evaluate available AI models for face and eye tracking. o Create a proof of concept for real-time video adjustment. • Phase 2: Backend Development (4 weeks) o Set up the Zoom Video SDK and APIs. o Implement video stream processing logic. • Phase 3: Frontend Development (3 weeks) o Build the user interface with controls for eye-contact correction. o Integrate video preview and toggles. • Phase 4: Testing & Optimization (2 weeks) o Test for performance, latency, and accuracy. o Optimize for different hardware configurations. • Phase 5: Deployment (1 week) o Package the app for Zoom App Marketplace and/or as a standalone virtual camera. ________________________________________ 5. Deliverables • Fully functional app integrated with Zoom or operating as a virtual camera. • User-friendly interface with customization options. • Documentation: o User guide. o Developer documentation for future enhancements. • Deployment assistance to the Zoom App Marketplace (if applicable). ________________________________________ Upwork Project Description Title: AI-Powered Eye Contact Correcting App for Zoom Description: We’re looking for a talented developer or team to build an innovative AI-powered app for Zoom that corrects eye contact in real-time. This app will leverage advanced computer vision to adjust video streams and simulate natural eye contact, even when the user is looking at different parts of the screen. Key Features: • Real-time face and eye tracking using AI. • Integration with Zoom via the Zoom SDK or functionality as a virtual camera. • User-friendly interface for toggles and customization. • Optimized video processing for low latency. • Privacy-focused design with local video processing. Technical Requirements: • Experience with Zoom SDKs (Zoom Video SDK or Meeting SDK). • Proficiency in AI/Computer Vision tools (OpenCV, MediaPipe, Dlib). • Expertise in Python (for AI) and JavaScript/Node.js (for frontend/backend development). • Familiarity with WebRTC for real-time video manipulation. • Strong understanding of UI/UX design for seamless user experiences. Deliverables: • Fully functional desktop application or Zoom-integrated app. • Clear user and developer documentation. • Assistance with deployment to the Zoom App Marketplace or packaging as a standalone virtual camera. Budget: $[Your Budget Range, e.g., $5,000-$10,000] (open to discussion based on expertise). Timeline: [Insert your preferred timeline, e.g., 10-12 weeks]. If you’re experienced in developing cutting-edge AI applications and want to be part of a groundbreaking project, we’d love to hear from you! Submit Your Proposal Today and join us in transforming video communication.

I need 1100+ m3u8 files to be converted to mp4 with specific names. Delivered to you are a CSV file with - URL's to the m3u8 files - UserID - First and last name of receiver - Name of the client manager So the task is to convert 1100+ video's from m3u8 to mp4 using ffmpeg or some other program and name the files correctly. The files should consist of the UserID, first+last name of the receiver and the client manager. Deliverables: The video's are short form, max 2.5 minutes, mostly shorter. You deliver a folder with all video's as mp4 files of max 25mb with the following naming. UserID_Name Receiver_Client Manager.mp4 Please only reply if you are able to do this on short notice and with a price for 1100+ video's. Thanks!

IF YOUR LOOKING FOR A 5* REVIEW TO BOOST YOUR PROFILE APPLY TO THIS JOB I need someone who can fix the code i have attached to the job the problems are in this loom https://www.loom.com/share/35d7bb77a58c4dfa8ab1a0ac240b589f?sid=49203439-796e-45af-a529-40b3ec615147

I have a hosted Python application using VPS and I need to solve an issue that it is not processing the video. I'm using MoviePy and FFmpeg. Will might need to check configurations or update/install packages. AnyDesk is required, Regards, Shoam

Hi, are you available for a quick SSLVPN issue the certificate is installed but for some reason it started acting like it’s expired

IF you have any experience with movie py and ffmpeg apply to this job i am after a freelancer who has worked in a project where you turn writing into video so only apply if you have this experience