PointCloud Saliency Maps

Tianhang Zheng, Changyou Chen, Junsong Yuan, Bo Li, Kui Ren; The IEEE International Conference on Computer Vision (ICCV), 2019, pp. 1598-1606


3D point-cloud recognition with PointNet and its variants has received remarkable progress. A missing ingredient, however, is the ability to automatically evaluate point-wise importance w.r.t. classification performance, which is usually reflected by a saliency map. A saliency map is an important tool as it allows one to perform further processes on point-cloud data. In this paper, we propose a novel way of characterizing critical points and segments to build point-cloud saliency maps. Our method assigns each point a score reflecting its contribution to the model-recognition loss. The saliency map explicitly explains which points are the key for model recognition. Furthermore, aggregations of highly-scored points indicate important segments/subsets in a point-cloud. Our motivation for constructing a saliency map is by point dropping, which is a non-differentiable operator. To overcome this issue, we approximate point-dropping with a differentiable procedure of shifting points towards the cloud centroid. Consequently, each saliency score can be efficiently measured by the corresponding gradient of the loss w.r.t the point under the spherical coordinates. Extensive evaluations on several state-of-the-art point-cloud recognition models, including PointNet, PointNet++ and DGCNN, demonstrate the veracity and generality of our proposed saliency map. Code for experiments is released on https://github.com/tianzheng4/PointCloud-Saliency-Maps

Related Material

author = {Zheng, Tianhang and Chen, Changyou and Yuan, Junsong and Li, Bo and Ren, Kui},
title = {PointCloud Saliency Maps},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}