Contact person: Real-Time AI Video Upscaler

Phone:Show

Email:Show

Location: Naogaon, Bangladesh

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"I’m looking to turn standard-definition MP4 footage into crisp HD (or better) in real time through an AI-powered application. The upscaled material must be broadcast-ready for television and equally polished for fast-moving social feeds, so latency has to stay low while color accuracy and sharpness remain high.

Core expectations
The program should ingest an SD MP4 stream or file, process it on-the-fly—ideally using GPU acceleration—and output a clean up-resolution signal that slots straight into a TV playout chain or can be rendered for social platforms. If you can keep the door open for other common containers down the line, that’s a welcome bonus, but MP4 is the non-negotiable starting point.

Quality & performance targets
• Real-time (or near real-time) processing at 25/30 fps for PAL/NTSC sources
• Minimal artifacting, with motion-aware enhancement to avoid the “over-sharpened” look
• Optional preset profiles tuned for broadcast versus mobile viewing
• Straightforward UI or command-line operation on Windows or Linux; I’m flexible here, so propose what best fits your approach.

Deliverables
1. Compiled application (or well-documented script) ready to install and run
2. Trained model files and clear instructions for re-training or fine-tuning
3. Brief user guide covering setup, sample commands, and recommended hardware
4. A short demo clip proving real-time upscaling from SD MP4 to HD.

Acceptance will be based on hitting the frame-rate target, visual quality benchmarks I’ll provide, and smooth playback on a broadcast monitor and a mobile device.

If you have experience with ESRGAN, Real-ESRGAN, or similar neural networks and can demonstrate live processing, let’s talk about timelines and milestones right away." (client-provided description)