MixPolyp: Integrating Mask, Box and Scribble Supervision for Enhanced Polyp Segmentation

TL;DR

MixPolyp introduces a multi-annotation supervised approach for polyp segmentation, combining mask, box, and scribble annotations with novel loss functions to improve performance and reduce labeling costs.

Contribution

It proposes a unified framework with three new supervision losses that handle diverse annotations, enhancing data efficiency without increasing inference complexity.

Findings

Outperforms existing methods on five datasets.

Reduces labeling costs by utilizing mixed supervision.

Effective across various annotation types and datasets.

Abstract

Limited by the expensive labeling, polyp segmentation models are plagued by data shortages. To tackle this, we propose the mixed supervised polyp segmentation paradigm (MixPolyp). Unlike traditional models relying on a single type of annotation, MixPolyp combines diverse annotation types (mask, box, and scribble) within a single model, thereby expanding the range of available data and reducing labeling costs. To achieve this, MixPolyp introduces three novel supervision losses to handle various annotations: Subspace Projection loss (L_SP), Binary Minimum Entropy loss (L_BME), and Linear Regularization loss (L_LR). For box annotations, L_SP eliminates shape inconsistencies between the prediction and the supervision. For scribble annotations, L_BME provides supervision for unlabeled pixels through minimum entropy constraint, thereby alleviating supervision sparsity. Furthermore, L_LR provides dense supervision by enforcing consistency among the predictions, thus reducing the non-uniqueness. These losses are independent of the model structure, making them generally applicable. They are used only during training, adding no computational cost during inference. Extensive experiments on five datasets demonstrate MixPolyp's effectiveness.