accessory
发表于 2025-4-1 05:55:52
http://reply.papertrans.cn/25/2424/242346/242346_61.png
CLASH
发表于 2025-4-1 09:17:49
http://reply.papertrans.cn/25/2424/242346/242346_62.png
PATHY
发表于 2025-4-1 11:52:15
,Adaptive Human Trajectory Prediction via Latent Corridors,rring trends from extremely small amounts of new data (e.g., 2 humans observed for 30 s). With less than . additional model parameters, we see up to . ADE improvement in MOTSynth simulated data and . ADE in MOT and Wildtrack real pedestrian data. Qualitatively, we observe that latent corridors imbue
Cuisine
发表于 2025-4-1 17:35:06
,Video Question Answering with Procedural Programs,rizons in order to answer complex questions. This code generation framework additionally enables ProViQ to perform other video tasks beyond question answering, such as multi-object tracking or basic video editing. ProViQ achieves state-of-the-art results on a diverse range of benchmarks, with improv
Medley
发表于 2025-4-1 21:00:20
http://reply.papertrans.cn/25/2424/242346/242346_65.png
处理
发表于 2025-4-2 01:13:17
,Tri,-plane: Thinking Head Avatar via Feature Pyramid,reconstruction of complex and dynamic head movements from monocular videos still suffers from capturing and restoring fine-grained details. In this work, we propose a novel approach, named Tri.-plane, for monocular photo-realistic volumetric head avatar reconstructions. Distinct from the existing wo