Shenzhen Afirstsoft Technology Group Co,.Ltd. China
send
badge
AI超高清视频修复增强系统
AI
QIONMAX AI UHD Video Restoration and Enhancement System is an innovative product developed by Soft Bull Technology based on self-developed AI algorithms. It strictly follows broadcast television ultra-high-definition technical standards and is suitable for multiple professional application scenarios such as broadcast-level program production, historical footage restoration, and cultural data digitization. The system deeply integrates cutting-edge artificial intelligence technologies and incorporates core algorithms including super-resolution reconstruction, intelligent denoising, compression artifact restoration, facial enhancement restoration, AI frame interpolation, and SDR to HDR conversion. By leveraging deep learning and generative computer vision models, it achieves comprehensive restoration and quality enhancement of common video issues such as blurriness, noise, scratches, and artifacts.

1. Supports visual restoration and production of HD/UHD program materials, supports various common video file formats, including MP4, MOV, MXF, MPEG2, AVC, etc.; color space support includes Rec. 709 and Rec. 2020; supports HLG/PQ; supports standard definition/HD, ultra HD 4K/8K resolutions; supports various common audio formats, and output files can retain the source file's audio format.

 

2. High-efficiency accelerated conversion, supports professional GPU computing power, supports 4K broadcast TV program standard format output.

 

3. Supports video playback on various scenarios from large screens to small screens, supports custom output video files at any resolution, and supports horizontal and vertical screen conversion.

 

4. Supports AI automatic detection and analysis of video issues and recommends relevant models, and users can manually toggle this feature.

 

5. Supports multiple AI video restoration and enhancement algorithms for visual production, including 4K/8K super-resolution, clarity enhancement, denoising, compression repair, detail enhancement, face enhancement, AI frame interpolation, SDR/HDR conversion, stabilization, etc. Algorithm models can be used individually or combined for multiple model effects; algorithm models support custom parameter adjustments.

 

6. Supports visual real-time comparison preview, supports zoomed preview, preview modes can be single video or dual video, and preview duration supports 5 frames, 3 seconds, 5 seconds, or 30 seconds.

 

7. Supports segmenting videos on the timeline and applying different algorithms to each segment.

 

8. Supports exporting queue files, allows pausing and resuming during export, supports multiple batch exports, and can display export progress in real-time.

More Products
Help Center

support_agent
Support


share
Social Share