Practical-RIFE — real-time frame interpolation

Description

Actively maintained, deployment-focused fork of RIFE (real-time intermediate flow estimation) for arbitrary-timestep video frame interpolation.

URL

• Visit resource

Related resources

• BackgroundMattingV2Real-time high-resolution background matting requiring a captured background plate, achieving 4K 30fps on RTX 2080 Ti, companion to arXiv:2…
• RobustVideoMattingReal-time human video matting via recurrent neural network requiring no green screen, achieving 4K 76FPS on GTX 1080 Ti, companion to arXiv…
• enhancrElectron GUI for TensorRT/NCNN frame interpolation and super-resolution, polished alternative to Flowframes.
• REAL-Video-EnhancerCross-platform GUI app for video interpolation/upscaling/denoising/decompression using TensorRT and NCNN backends.
• ComfyUI-VideoColorGradingComfyUI node pack for video color grading, integrating into node-based AI compositing workflows.
• Evaluating Video Quality Metrics for Neural and Traditional Codecs (4K/UHD-1)Academic comparison of VMAF, AVQBits, and FasterVQA no-reference quality metrics on 4K content.
• Bitrate Ladder Construction via Transfer Learning and Spatio-Temporal FeaturesML-based bitrate ladder optimization achieving 94.1% complexity reduction at only 1.71% BD-rate cost using transfer learning.
• VMAF-torchPyTorch reimplementation of VMAF metric, GPU/gradient-friendly for use in learned codec optimization loops.
• CompressedVQA: Full/No-Reference VQA for Compressed UGC VideoICME 2021 grand challenge-winning full-reference and no-reference video quality models for compressed user-generated content.
• 2BiVQA: Double Bi-LSTM No-Reference Video Quality AssessmentNo-reference VQA model for UGC video using double Bi-LSTM, companion code to arXiv:2208.14774.