Cascade Corner Pooling

Introduced by Duan et al. in CenterNet: Keypoint Triplets for Object Detection

Cascade Corner Pooling is a pooling layer for object detection that builds upon the corner pooling operation. Corners are often outside the objects, which lacks local appearance features. CornerNet uses corner pooling to address this issue, where we find the maximum values on the boundary directions so as to determine corners. However, it makes corners sensitive to the edges. To address this problem, we need to let corners see the visual patterns of objects. Cascade corner pooling first looks along a boundary to find a boundary maximum value, then looks inside along the location of the boundary maximum value to find an internal maximum value, and finally, add the two maximum values together. By doing this, the corners obtain both the the boundary information and the visual patterns of objects.

Source: CenterNet: Keypoint Triplets for Object Detection

Latest Papers

Efficient grouping for keypoint detection
Alexey SidnevEkaterina KrasikovaMaxim Kazakov
DPAttack: Diffused Patch Attacks against Universal Object Detection
Shudeng WuTao DaiShu-Tao Xia
Non-anchor-based vehicle detection for traffic surveillance using bounding ellipses
Byeonghyeop YuJohyun ShinGyeongjun KimSeungbin RohKeemin Sohn
Object Detection and Tracking Algorithms for Vehicle Counting: A Comparative Analysis
Vishal MandalYaw Adu-Gyamfi
DeepMark++: CenterNet-based Clothing Detection
Alexey SidnevAlexander KrapivinAlexey TrushkovEkaterina KrasikovaMaxim Kazakov
Deep Active Learning for Remote Sensing Object Detection
Zhenshen QuJingda DuYong CaoQiuyu GuanPengbo Zhao
DeepMark: One-Shot Clothing Detection
Alexey SidnevAlexey TrushkovMaxim KazakovIvan KorolevVladislav Sorokin
Enhancing 2D Representation via Adjacent Views for 3D Shape Retrieval
Cheng Xu Zhaoqun Li Qiang Qiu Biao Leng Jingfei Jiang
Object detection on aerial imagery using CenterNet
Dheeraj Reddy PaillaVarghese KollerathuSai Saketh Chennamsetty
Deep High-Resolution Representation Learning for Visual Recognition
| Jingdong WangKe SunTianheng ChengBorui JiangChaorui DengYang ZhaoDong LiuYadong MuMingkui TanXinggang WangWenyu LiuBin Xiao
CenterNet: Keypoint Triplets for Object Detection
| Kaiwen DuanSong BaiLingxi XieHonggang QiQingming HuangQi Tian