End-to-End Weakly Supervised Semantic Segmentation with Co-training and Swapping Assignments
The author argues that by optimizing CAMs in an end-to-end manner, the reliance on refinement processes can be reduced, leading to more reliable and accurate CAMs for weakly supervised semantic segmentation. The proposed method, Co-training with Swapping Assignments (CoSA), leverages a dual-stream framework to achieve exceptional performance.
Auxiliary Tasks Enhanced Dual-affinity Learning for Weakly Supervised Semantic Segmentation
The author proposes an effective multi-task auxiliary deep learning framework for weakly supervised semantic segmentation, leveraging saliency detection and image classification as auxiliary tasks to aid the primary task. They introduce a cross-task dual-affinity learning module to refine both pairwise and unary affinities, enhancing task-specific features and predictions.
弱教師付き意味論的セグメンテーションのためのコンテキスト・プロトタイプ認識学習
コンテキストの知識バイアスが、インスタンスの意味的特徴を十分に理解する能力に影響を与えることを指摘し、コンテキスト・プロトタイプ認識学習を提案する。これにより、インスタンスの多様な属性を適応的に捉えることができ、より正確で完全なクラス活性化マップを生成できる。
Co-training with Swapping Assignments for Semantic Segmentation
End-to-end model CoSA improves CAM consistency and accuracy, outperforming existing methods.
Weakly Supervised Semantic Segmentation: Tackling Co-occurrence with SeCo
SeCo proposes a 'Separate and Conquer' scheme to address the co-occurrence issue in weakly supervised semantic segmentation.
Reducing Background Noise in Attention Maps for Improved Weakly Supervised Semantic Segmentation
The proposed method reduces background noise in attention maps by incorporating attention map-enhanced class activation maps into the loss function during training, leading to improved segmentation accuracy.
© 2024 by Linnk AI