On the heels of a computer vision system that achieved state-of-the-art accuracy with minimal supervision, Facebook today announced a project called Learning from Videos that's designed to ...
Facebook AI 提出新型视频理解架构：完全基于Transformer，无需卷积，训练速度快、计算成本低。 TimeSformer 是首个完全基于 Transformer 的视频架构。近年来，Transformer 已成为自然语言处理（NLP）领域中 ...
TimeSformer 在几个难度系数比较高的动作识别基准上获得了最好的效果，包括 Kinetics-400动作识别数据集。此外，与现代3D 卷积神经网络(CNN)相比，TimeSformer 的训练速度提高了大约3倍，推理所需计算 ...
Results that may be inaccessible to you are currently showing.
Hide inaccessible results