High-resolution video matting requires preserving fine details, such as hair strands and soft transparency, while maintaining consistency across frames. Memory-based methods achieve strong matting qua
MatAnyone-E accelerates high-resolution video matting by selectively applying high-resolution matching to uncertain regions and using a lightweight decoder, achieving large speedups with preserved quality.
High-resolution video matting requires preserving fine details, such as hair strands and soft transparency, while maintaining consistency across frames. Memory-based methods achieve strong matting quality by matching current-frame features with information stored from earlier frames, but this process becomes increasingly costly as resolution grows. We present MatAnyone-E, an efficient extension of MatAnyone 2 that focuses computation where it is most needed. Our approach first performs low-resolution propagation to identify uncertain regions that require closer inspection and the memory candidates most relevant to them. It then performs high-resolution matching only for these regions using…
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
MatAnyone-E: Sparse Memory Propagation for Efficient High-Resolution Video Matting was first seen on Hugging Face · Daily papers on Oct 28, 2026. Since then it has appeared in 23 checks on 2 different days. Its best position was #12 of 30, reached on Oct 28, 05:28 UTC. It dropped off the list after Oct 29, 23:28 UTC.
Position on Hugging Face · Daily papers at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hugging Face.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Oct 29, 23:28 UTC, after 2 days of trending.
#12 on Hugging Face · Daily papers, on Oct 28, 05:28 UTC. Trendoday checks this list every 120 minutes.
It was first seen on Oct 28, 2026 and has been on the list on 2 days (23 checks) since.