G-CAME: Gaussian-Class Activation Mapping Explainer for Object Detectors
This provides interpretability for users of object detection models, but it is incremental as it builds on existing CAM-based approaches.
The authors tackled the problem of explaining object detection models by proposing G-CAME, a method that generates saliency maps using activation maps and Gaussian kernels, and evaluated it on YOLOX with the MS-COCO 2017 dataset, showing it is fast compared to other region-based methods.
Nowadays, deep neural networks for object detection in images are very prevalent. However, due to the complexity of these networks, users find it hard to understand why these objects are detected by models. We proposed Gaussian Class Activation Mapping Explainer (G-CAME), which generates a saliency map as the explanation for object detection models. G-CAME can be considered a CAM-based method that uses the activation maps of selected layers combined with the Gaussian kernel to highlight the important regions in the image for the predicted box. Compared with other Region-based methods, G-CAME can transcend time constraints as it takes a very short time to explain an object. We also evaluated our method qualitatively and quantitatively with YOLOX on the MS-COCO 2017 dataset and guided to apply G-CAME into the two-stage Faster-RCNN model.