Saliency-Guided Perceptual Grouping Using Motion Cues in Region-Based Artificial Visual Attention
This work addresses a domain-specific need in computer vision and robotics for more effective object grouping after attention processes, but it appears incremental as it builds on existing region-based frameworks.
The paper tackles the problem of grouping segmented regions into objects for post-attentional tasks like identification or tracking, by proposing a saliency-guided approach that uses proximity and motion similarity, and compares it to other methods.
Region-based artificial attention constitutes a framework for bio-inspired attentional processes on an intermediate abstraction level for the use in computer vision and mobile robotics. Segmentation algorithms produce regions of coherently colored pixels. These serve as proto-objects on which the attentional processes determine image portions of relevance. A single region---which not necessarily represents a full object---constitutes the focus of attention. For many post-attentional tasks, however, such as identifying or tracking objects, single segments are not sufficient. Here, we present a saliency-guided approach that groups regions that potentially belong to the same object based on proximity and similarity of motion. We compare our results to object selection by thresholding saliency maps and a further attention-guided strategy.