2020/03/31 by Shang-Hua Gao, Yong-Qiang Tan, Ming-Ming Cheng +3 · 1 citation
Computer Science · #cs.CV
published as ECCV 2020 · Accepted by ECCV 2020. Source code: https://mmcheng.net/sod100k/
arxiv created 2020/08/02 · arxiv updated 2020/08/04
Salient object detection models often demand a considerable amount of computation cost to make precise prediction for each pixel, making them hardly applicable on low-power devices. In this paper, we aim to relieve the contradiction between computation cost and model performance by improving the network efficiency to a higher degree. We propose a flexible convolutional module, namely generalized OctConv (gOctConv), to efficiently utilize both in-stage and cross-stages multi-scale features, while reducing the representation redundancy by a novel dynamic weight decay scheme. The effective dynamic weight decay scheme stably boosts the sparsity of parameters during training, supports learnable number of channels for each scale in gOctConv, allowing 80% of parameters reduce with negligible performance drop. Utilizing gOctConv, we build an extremely light-weighted model, namely CSNet, which achieves comparable performance with about 0.2% parameters (100k) of large models on popular salient object detection benchmarks.