DailyPapers (@HuggingPapers)
2026-01-06 | โค๏ธ 19 | ๐ 2
AdaGaR
A unified framework for dynamic 3D scene reconstruction from monocular videos.
Uses Adaptive Gabor Representation to capture high-frequency details and Cubic Hermite Splines for temporal continuity.
Achieves SOTA on Tap-Vid DAVIS with PSNR 35.49. https://x.com/HuggingPapers/status/2008333133116998043/photo/1
๐ ์๋ณธ ๋งํฌ
๋ฏธ๋์ด
๐ Related
Auto-generated - needs manual review