Timezone: »
Soft-argmax operation is commonly adopted in detection-based methods to localize the target position in a differentiable manner. However, training the neural network with soft-argmax makes the shape of the probability map unconstrained. Consequently, the model lacks pixel-wise supervision through the map during training, leading to performance degradation. In this work, we propose sampling-argmax, a differentiable training method that imposes implicit constraints to the shape of the probability map by minimizing the expectation of the localization error. To approximate the expectation, we introduce a continuous formulation of the output distribution and develop a differentiable sampling process. The expectation can be approximated by calculating the average error of all samples drawn from the output distribution. We show that sampling-argmax can seamlessly replace the conventional soft-argmax operation on various localization tasks. Comprehensive experiments demonstrate the effectiveness and flexibility of the proposed method. Code is available at https://github.com/Jeff-sjtu/sampling-argmax
Author Information
Jiefeng Li (Shanghai Jiao Tong University)
Tong Chen (Shanghai Jiao Tong University)
Ruiqi Shi (Shanghai Jiao Tong University)
Yujing Lou (Shanghai Jiao Tong University)
Yong-Lu Li (Shanghai Jiao Tong University)
Cewu Lu (Shanghai Jiao Tong University)
More from the Same Authors
-
2022 Poster: DART: Articulated Hand Model with Diverse Accessories and Rich Textures »
Daiheng Gao · Yuliang Xiu · Kailin Li · Lixin Yang · Feng Wang · Peng Zhang · Bang Zhang · Cewu Lu · Ping Tan -
2020 Poster: HOI Analysis: Integrating and Decomposing Human-Object Interaction »
Yong-Lu Li · Xinpeng Liu · Xiaoqian Wu · Yizhuo Li · Cewu Lu