Research map / Detection and segmentation / CNN features
Hierarchical Cellular Automata for Visual Saliency
Yao Qin, Mengyang Feng, Huchuan Lu, Garrison W. Cottrell
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| combine pool1 and pool5 FCN deep features instead of a single layer in the impact factor (ablation) | Single-layer Cellular Automata · ECSSD (validation) | F-measure and MAE | Improved |
| add Single-layer Cellular Automata optimization to existing saliency maps (SCA200) | existing saliency methods (MCDL, MDF, LEGS, SSD-HS, etc.) · ECSSD and PASCAL-S | MAE and precision-recall | Improved |
| add multi-scale integration via CCA on top of SCA (HCA*) (ablation) | SCA · ECSSD and PASCAL-S | MAE | Improved |
| integrate three methods with Cuboid Cellular Automata | BL, HS, MR and DS, MCDL, MDF · ECSSD | precision, recall, F-measure, MAE | Improved |
| use CCA with adjacent cells across layers instead of MCA (ablation) | saliency map integration · ECSSD and MSRA | PR, F-measure, MAE | Improved |
| add SCA cleanup before integration (HCA instead of CCA/MCA) (ablation) | integration of FT, ITTI, CAS · ECSSD and MSRA | PR, F-measure, MAE | Improved |
| include SLIC superpixel segmentation in run time | HCA · MSRA5000 | run time (s): .2421 -> 1.0240 | Hurt |
| vary coherence constant b in [0.1, 0.3] with a fixed at 0.6 (ablation) | SCA coherence matrix · saliency detection | saliency quality | No clear effect |
Closest papers by meaning
- Attention-based Neural Cellular Automata (NeurIPS 2022 · similarity 0.87)
- Depth-Induced Multi-Scale Recurrent Attention Network for Saliency Detection (ICCV 2019 · similarity 0.86)
- Hierarchical Image Saliency Detection on Extended CSSD (TPAMI 2016 · similarity 0.86)
- Unsupervised CNN-based Co-Saliency Detection with Graphical Optimization (ECCV 2018 · similarity 0.86)
- Salient Object Detection with Recurrent Fully Convolutional Networks (TPAMI 2018 · similarity 0.86)
- DHSNet: Deep Hierarchical Saliency Network for Salient Object Detection (CVPR 2016 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
