Leon_yixiao的博客

Main Contributions: sub-region attention map and aspect ratio attention map for each RoI   RoI feature extractor adopt a 1 × 1 convo- lutional layer to reduce the channel number to Cs and pool...

Detection with Enriched Semantics (DES) is a singleshot object detection network with three parts a single shot detection branch, ssd a segmentation branch to enrich semantics at low level dete...

Contritutions: study the impact of crowd occlusion on pedestrian detection Two types of repulsion losses: RepGT Loss and RepBox Loss a crowd-robust pedestrian detector is trained end-to-end Lo...

where xt, xb, xl , xr represent the distances between curerent pixel location (i, j) and the top, bottom, left and right bounds of ground truth, respectively L2 Loss Layer: two drawbacks...