TimeSformer在K600上预训练好的的模型:16 of frames,spatial crop:448,acc@1:81.8,acc@5:95.8。
TimeSformer:Is Space-Time Attention All You Need for Video Understanding?(video transformer)
TimeSformer:Is Space-Time Attention All You Need for Video Understanding?(video transformer)
TimeSformer在K400上预训练好的的模型:8 of frames,spatial crop:224,acc@1:77.9,acc@5:93.2。