NAS-Count: Counting-by-Density with Neural Architecture Search
arXiv:2003.00217
Abstract
Most of the recent advances in crowd counting have evolved from hand-designed density estimation networks, where multi-scale features are leveraged to address the scale variation problem, but at the expense of demanding design efforts. In this work, we automate the design of counting models with Neural Architecture Search (NAS) and introduce an end-to-end searched encoder-decoder architecture, Automatic Multi-Scale Network (AMSNet). Specifically, we utilize a counting-specific two-level search space. The encoder and decoder in AMSNet are composed of different cells discovered from micro-level search, while the multi-path architecture is explored through macro-level search. To solve the pixel-level isolation issue in MSE loss, AMSNet is optimized with an auto-searched Scale Pyramid Pooling Loss (SPPLoss) that supervises the multi-scale structural information. Extensive experiments on four datasets show AMSNet produces state-of-the-art results that outperform hand-designed models, fully demonstrating the efficacy of NAS-Count.
arXiv commentAccepted to European Conference on Computer Vision(ECCV) 2020
References in corpus (26)
- Neural Architecture Search with Reinforcement Learning
- Geometric deep learning: going beyond Euclidean data
- Neural Architecture Search: A Survey
- A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimation
- Efficient Neural Architecture Search via Parameter Sharing
- Large-Scale Evolution of Image Classifiers
- PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search
- SMASH: One-Shot Model Architecture Search through HyperNetworks
- Searching for Efficient Multi-Scale Architectures for Dense Image Prediction
- Efficient Architecture Search by Network Transformation
- SNAS: Stochastic Neural Architecture Search
- DARTS+: Improved Differentiable Architecture Search with Early Stopping
- Hierarchical Representations for Efficient Architecture Search
- The Evolved Transformer
- Simple And Efficient Architecture Search for Convolutional Neural Networks
- DetNAS: Backbone Search for Object Detection
- Convolutional Neural Fabrics
- Progressive Differentiable Architecture Search: Bridging the Depth Gap between Search and Evaluation
- Light-Weight RefineNet for Real-Time Semantic Segmentation
- Exploring Randomly Wired Neural Networks for Image Recognition
- Crowd Counting and Density Estimation by Trellis Encoder-Decoder Network
- Mixture of Counting CNNs: Adaptive Integration of CNNs Specialized to Specific Appearance for Crowd Counting
- Composition Loss for Counting, Density Map Estimation and Localization in Dense Crowds
- Stacked Pooling: Improving Crowd Counting by Boosting Scale Invariance
- Leveraging Unlabeled Data for Crowd Counting by Learning to Rank
- In Defense of Single-column Networks for Crowd Counting