Highly Efficient Salient Object Detection with 100K Parameters

Abstract

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.

arXiv commentAccepted by ECCV 2020. Source code: https://mmcheng.net/sod100k/

References in corpus (15)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
• Res2Net: A New Multi-scale Backbone Architecture
• ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
• Deeply supervised salient object detection with short connections
• DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detection
• Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network
• DABNet: Depth-wise Asymmetric Bottleneck for Real-time Semantic Segmentation
• Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution
• MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning
• CGNet: A Light-weight Context Guided Network for Semantic Segmentation
• Towards High-Resolution Salient Object Detection
• Three Birds One Stone: A General Architecture for Salient Object Segmentation, Edge Detection and Skeleton Extraction
• Linear Context Transform Block

Cited by in corpus (3)

• Uncertainty-aware Joint Salient Object and Camouflaged Object Detection
• AdversarialNAS: Adversarial Neural Architecture Search for GANs
• One-Shot Object Affordance Detection in the Wild