A ResNet block that fuses channel-wise temporal convolution with learned feature-level motion subtraction achieves state-of-the-art accuracy on temporal video benchmarks.
Flownet: Learning optical flow with convolutional networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
STM: SpatioTemporal and Motion Encoding for Action Recognition
A ResNet block that fuses channel-wise temporal convolution with learned feature-level motion subtraction achieves state-of-the-art accuracy on temporal video benchmarks.