PT Unknown
AU Abel Gonzalez-Garcia
   Davide Modolo
   Vittorio Ferrari
TI Objects as context for detecting their semantic parts
BT 31st IEEE Conference on Computer Vision and Pattern Recognition
PY 2018
BP 6907
EP 6916
DI 10.1109/CVPR.2018.00722
DE Proposals; Semantics; Wheels; Automobiles; Context modeling; Task analysis; Object detection
AB We present a semantic part detection approach that effectively leverages object information. We use the object appearance and its class as indicators of what parts to expect. We also model the expected relative location of parts inside the objects based on their appearance. We achieve this with a new network module, called OffsetNet, that efficiently predicts a variable number of part locations within a given object. Our model incorporates all these cues todetect parts in the context of their objects. This leads to considerably higher performance for the challenging task of part detection compared to using part appearance alone (+5 mAP on the PASCAL-Part dataset). We also compareto other part detection methods on both PASCAL-Part and CUB200-2011 datasets.
ER