Search papers, labs, and topics across Lattice.
This paper introduces a novel framework that enhances weakly supervised volumetric segmentation by rectifying noisy class activation maps (CAMs) without additional training or prototypes. By leveraging temporal and structural coherence in volumetric data, the authors present Variance-Reduced Activation Aggregation (VRAA) to suppress noise and amplify coherent signals, alongside a Bidirectional Extremity Rectification (BER) mechanism to correct implausible activations. The proposed method achieves significant performance improvements on multiple benchmarks, with up to 20% gains in Dice and 40% in mIoU, while also reducing inference time by over five times compared to existing state-of-the-art methods.
Exploiting temporal coherence allows for a dramatic reduction in noise and a fivefold increase in inference efficiency for weakly supervised segmentation.
Weakly supervised segmentation relies heavily on class activation maps (CAMs) to initially localize target regions. However, CAMs are often noisy and prone to catastrophic failures. Existing remedies typically introduce additional training stages or prototype learning, increasing computational cost and reducing robustness. In this paper, we propose a training-free prototype-free framework that rectifies unreliable CAMs by exploiting temporal and structural coherence in volumetric data as a free lunch. Our approach is built on two key components. First, we introduce Variance-Reduced Activation Aggregation (VRAA) which suppresses noise and amplify coherent semantic signals. We provide a theoretical justification by modeling CAMs as high-dimensional random vectors and show that aggregation yields provable variance reduction. Second, we design a Bidirectional Extremity Rectification (BER) mechanism that detects and rectifies implausible activations through bidirectional extremity checks, effectively mitigating extreme-value failures without learning additional parameters. Our method is model-agnostic and can be seamlessly integrated with existing pipelines. Extensive experiments on multiple public benchmarks demonstrate substantial improvements over state-of-the-art weakly supervised methods, achieving up to 20% Dice and 40% mIoU gains while reducing inference time by more than 5 times. These results indicate that leveraging coherence as an implicit inductive bias yields a principled and efficient approach to stabilizing weakly supervised volumetric segmentation. Our code will be available.
