Revisiting AP Loss for Dense Object Detection: Adaptive Ranking Pair Selection
This work addresses a specific bottleneck in object detection for computer vision researchers, offering incremental improvements in ranking loss design.
The authors tackled the problem of improving dense object detection by revisiting the Average Precision (AP) loss, focusing on selecting ranking pairs between positive and negative samples, and proposed an Adaptive Pairwise Error (APE) loss and a clustering-based strategy, achieving superior performance on the MSCOCO dataset compared to current methods.
Average precision (AP) loss has recently shown promising performance on the dense object detection task. However,a deep understanding of how AP loss affects the detector from a pairwise ranking perspective has not yet been developed.In this work, we revisit the average precision (AP)loss and reveal that the crucial element is that of selecting the ranking pairs between positive and negative samples.Based on this observation, we propose two strategies to improve the AP loss. The first of these is a novel Adaptive Pairwise Error (APE) loss that focusing on ranking pairs in both positive and negative samples. Moreover,we select more accurate ranking pairs by exploiting the normalized ranking scores and localization scores with a clustering algorithm. Experiments conducted on the MSCOCO dataset support our analysis and demonstrate the superiority of our proposed method compared with current classification and ranking loss. The code is available at https://github.com/Xudangliatiger/APE-Loss.