LSTM Pose Machines

CVPR 2018 Yue LuoJimmy RenZhouxia WangWenxiu SunJinshan PanJianbo LiuJiahao PangLiang Lin

We observed that recent state-of-the-art results on single image human pose estimation were achieved by multi-stage Convolution Neural Networks (CNN). Notwithstanding the superior performance on static images, the application of these models on videos is not only computationally intensive, it also suffers from performance degeneration and flicking... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Pose Estimation J-HMDB LSTM PM Mean [email protected] 93.6 # 1
Pose Estimation UPenn Action LSTM PM Mean [email protected] 97.7 # 1

Methods used in the Paper