A local attention transformer used as the final upsampler reduces optical flow endpoint error on Sintel Clean for RAFT, GMA, and FlowFormer.
Training stochastic model recognition algo- rithms as networks can lead to maximum mutual information estimation of parameters
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Local Attention Transformers for High-Detail Optical Flow Upsampling
A local attention transformer used as the final upsampler reduces optical flow endpoint error on Sintel Clean for RAFT, GMA, and FlowFormer.