Lin et al (2017) Focal Loss for Dense Object Detection

March 15, 2023 paper

This is the paper proposed RetinaNet and also the focal loss function to better train object detection models. Object detection models are in two camps, there are two-stage proposal-driven models such as R-CNN, and one-stage detector such that YOLO and SSD. The paper claimed that the prior result on one-stage... [more]

Lin et al (2017) Feature Pyramid Networks for Object Detection

March 3, 2023 paper

This paper proposed feature pyramid network to find scale invariant object detection, i.e., a model that can detect objects of different scales. One way to tackle scale invariant problem is to form an image pyramid of different scale and process each with the same model. This is a brute-force approach.... [more]

He, Gkioxari and Dollar and Girshick (2017) Mask R-CNN

March 3, 2023 paper

This is the Mask R-CNN paper. It extends Faster R-CNN to produce pixel masks, i.e., for semantic segmentation. It extends Faster R-CNN by adding a branch for predicting segmentation masks on each Region of Interest (RoI). However, this have to support with RoIAlign instead of RoIPooling, to avoid error introduced... [more]