FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

CVPR 2017 Eddy IlgNikolaus MayerTonmoy SaikiaMargret KeuperAlexey DosovitskiyThomas Brox

The FlowNet demonstrated that optical flow estimation can be cast as a learning problem. However, the state of the art with regard to the quality of the flow has still been defined by traditional methods... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Dense Pixel Correspondence Estimation HPatches FlowNet2 Viewpoint I AEPE 5.99 # 4
Viewpoint II AEPE 15.55 # 4
Viewpoint III AEPE 17.09 # 4
Viewpoint IV AEPE 22.13 # 4
Viewpoint V AEPE 30.68 # 4
Skeleton Based Action Recognition JHMDB Pose Tracking FlowNet2 [email protected] 45.2 # 2
[email protected] 62.9 # 3
[email protected] 73.5 # 3
[email protected] 80.6 # 3
[email protected] 85.5 # 3
Optical Flow Estimation Sintel-clean FlowNet2 Average End-Point Error 3.96 # 9

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet