FILM: Frame Interpolation for Large Motion

Google Research's unified neural network for video frame interpolation handling large motion without requiring optical flow or depth estimation.

Link

• Visit resource

Related resources

• RIFE (ECCV2022)Real-time flow-based video frame interpolation model, widely used as the backend for many interpolation tools.
• Track-AnythingInteractive video object tracking and segmentation tool built on Segment Anything, XMem, and E2FGVI, feeding video inpainting pipelines.
• ProPainterICCV 2023 video inpainting tool for object removal/completion using propagation and transformer-based approach, CLI-usable.
• videoparser-ngCLI/API video bitstream parser (H.264/HEVC/VP9) extracting QP/motion/block-size stats without full decode, research-grade.
• MoMoAAAI 2025 "Disentangled Motion Modeling for Video Frame Interpolation," diffusion-based motion disentanglement approach.
• PXNR No-Reference Video Quality EstimationEricsson's sample implementation of the ITU-T P.NATS phase 2 no-reference video quality estimation module.
• ffmpeg-queueFastify + Redis job queue for dispatching FFmpeg transcoding jobs, Docker Compose ready, production-used.
• fssimu2Zig implementation of SSIMULACRA2 metric, ~23% faster and ~40% less memory than the reference implementation.
• BVQA_BenchmarkBlind (no-reference) VQA benchmarking toolkit comparing multiple models.
• CompressedVQA-AEVFull-reference + no-reference VQA for asymmetric-encoded video, QoMEX 2026 challenge solution using SigLIP2+Swin-B.