Description
RepPoints is a representation for object detection that consists of a set of points which indicate the spatial extent of an object and semantically significant local areas. This representation is learned via weak localization supervision from rectangular ground-truth boxes and implicit recognition feedback. Based on the richer RepPoints representation, the authors develop an anchor-free object detector that yields improved performance compared to using bounding boxes.
Papers Using This Method
Dual Scale-aware Adaptive Masked Knowledge Distillation for Object Detection2025-01-13CNN based Cuneiform Sign Detection Learned from Annotated 3D Renderings and Mapped Photographs with Illumination Augmentation2023-08-22AMD: Adaptive Masked Distillation for Object Detection2023-01-31Sample hardness based gradient loss for long-tailed cervical cell detection2022-08-07Point RCNN: An Angle-Free Framework for Rotated Object Detection2022-05-28Focal and Global Knowledge Distillation for Detectors2021-11-23Oriented RepPoints for Aerial Object Detection2021-05-24RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder2020-10-29RepPoints V2: Verification Meets Regression for Object Detection2020-07-16Dense RepPoints: Representing Visual Objects with Dense Point Sets2019-12-24RepPoints: Point Set Representation for Object Detection2019-04-25