Abstract:Saliency detection, finding the most important parts of an image, has become increasingly popular in computer vision. In this paper, we introduce Hierarchical Cellular Automata (HCA) – a temporally evolving model to intelligently detect salient objects. HCA consists of two main components: Single-layer Cellular Automata (SCA) and Cuboid Cellular Automata (CCA). As an unsupervised propagation mechanism, Single-layer Cellular Automata can exploit the intrinsic relevance of similar regions through interactions with neighbors. Low-level image features as well as highlevel semantic information extracted from deep neural networks are incorporated into the SCA to measure the correlation between different image patches. With these hierarchical deep features, an impact factor matrix and a coherence matrix are constructed to balance the influences on each cell’s next state. The saliency values of all cells are iteratively updated according to a well-defined update rule. Furthermore, we propose CCA to integrate multiple saliency maps generated by SCA at different scales in a Bayesian framework. Therefore, single-layer propagation and multi-scale integration are jointly modeled in our unified HCA. Surprisingly, we Yao Qin University of California, San Diego E-mail: yaq007@eng.ucsd.edu Mengyang Feng Dalian University of Technology E-mail: mengyangfeng@gmail.com Huchuan Lu Dalian University of Technology E-mail: lhchuan@dlut.edu.cn Garrsion W. Cottrell University of California, San Diego E-mail: gary@eng.ucsd.edu * Equal Contribution find that the SCA can improve all existing methods that we applied it to, resulting in a similar precision level regardless of the original results. The CCA can act as an efficient pixel-wise aggregation algorithm that can integrate state-of-the-art methods, resulting in even better results. Extensive experiments on four challenging datasets demonstrate that the proposed algorithm outperforms state-of-the-art conventional methods and is competitive with deep learning based approaches.

Segmentation Results Stimuli Final Saliency Map Ground Truth Constructed Graph CCA

Hierarchical Cellular Automata for Visual Saliency

Salient Region Detection Algorithm Via KL Divergence and Multi-scale Merging

Efficient Classification Using Salient Regions

Attention Shift-Based Multiple Saliency Object Segmentation.

Hierarchical Saliency Mapping for Weakly Supervised Object Localization Based on Class Activation Mapping

Saliency Detection Via the Improved Hierarchical Principal Component Analysis Method

Accurate salient object detection via dense recurrent connections and residual-based hierarchical feature integration.

Saliency Detection Based on Multiple-Level Feature Learning

CSA-Net: Deep Cross-Complementary Self Attention and Modality-Specific Preservation for Saliency Detection

Saliency Detection Via Global-Object-Seed-Guided Cellular Automata

Co-saliency Detection Based on Hierarchical Consistency.

Saliency-CCE: Exploiting colour contextual extractor and saliency-based biomedical image segmentation

Salient Object Detection and Segmentation

Towards accurate RGB-D saliency detection with complementary attention and adaptive integration

CAA : Channelized Axial Attention for Semantic Segmentation.

Co-saliency Detection Based on Superpixel Matching and Cellular Automata.

Saliency Detection by Selective Strategy for Salient Object Segmentation.

Hierarchical Contour Closure-Based Holistic Salient Object Detection

DeepACG: Co-Saliency Detection via Semantic-aware Contrast Gromov-Wasserstein Distance

SCSA: Exploring the Synergistic Effects Between Spatial and Channel Attention