Abstract:Weakly supervised semantic segmentation is receiving great attention due to its low human annotation cost. In this paper, we aim to tackle bounding box supervised semantic segmentation, i.e., training accurate semantic segmentation models using bounding box annotations as supervision. To this end, we propose Affinity Attention Graph Neural Network ($A^2$GNN). Following previous practices, we first generate pseudo semantic-aware seeds, which are then formed into semantic graphs based on our newly proposed affinity Convolutional Neural Network (CNN). Then the built graphs are input to our $A^2$GNN, in which an affinity attention layer is designed to acquire the short- and long- distance information from soft graph edges to accurately propagate semantic labels from the confident seeds to the unlabeled pixels. However, to guarantee the precision of the seeds, we only adopt a limited number of confident pixel seed labels for $A^2$GNN, which may lead to insufficient supervision for training. To alleviate this issue, we further introduce a new loss function and a consistency-checking mechanism to leverage the bounding box constraint, so that more reliable guidance can be included for the model optimization. Experiments show that our approach achieves new state-of-the-art performances on Pascal VOC 2012 datasets (val: 76.5\%, test: 75.2\%). More importantly, our approach can be readily applied to bounding box supervised instance segmentation task or other weakly supervised semantic segmentation tasks, with state-of-the-art or comparable performance among almot all weakly supervised tasks on PASCAL VOC or COCO dataset. Our source code will be available at <a class="link-external link-https" href="https://github.com/zbf1991/A2GNN" rel="external noopener nofollow">this https URL</a>.

An Adaptive Affinity Graph with Subspace Pursuit for Natural Image Segmentation

Affinity Fusion Graph-Based Framework for Natural Image Segmentation

Semantic Segmentation Using Multiple Graphs With Block-Diagonal Constraints

Adaptive hypergraph superpixels

Sparse Graph Connectivity for Image Segmentation

Adaptive fusion with multi-scale features for interactive image segmentation

Affinity Derivation for Accurate Instance Segmentation.

A Graph Segmentation Method Based on Compatibility Subgraph Aggregation

Label-efficient Segmentation via Affinity Propagation

Adaptive High-Precision Superpixel Segmentation

Graph-based Image Segmentation Using Directional Nearest Neighbor Graph

Affinity Derivation and Graph Merge for Instance Segmentation.

Revisiting graph construction for fast image segmentation

Affinity Attention Graph Neural Network for Weakly Supervised Semantic Segmentation

Efficient Region-Aware Large Graph Construction Towards Scalable Multi-Label Propagation

Affinity Space Adaptation for Semantic Segmentation Across Domains

Deep Affinity Net: Instance Segmentation via Affinity

Joint Dense-Point Representation for Contour-Aware Graph Segmentation

Automatic Fuzzy Clustering Framework for Image Segmentation

Learning Hybrid Semantic Affinity for Point Cloud Segmentation

Joint learning framework of superpixel generation and fuzzy sparse subspace clustering for color image segmentation