PCoTTA: Continual Test-Time Adaptation for Multi-Task Point Cloud Understanding
This addresses the problem of adapting models to evolving real-world environments for applications like autonomous driving or robotics, though it appears incremental as it builds on existing continual adaptation concepts.
The paper tackles continual test-time adaptation for multi-task point cloud understanding by proposing PCoTTA, a framework that handles multiple tasks in a unified model during adaptation, resulting in improved transferability to changing target domains as demonstrated through experimental comparisons.
In this paper, we present PCoTTA, an innovative, pioneering framework for Continual Test-Time Adaptation (CoTTA) in multi-task point cloud understanding, enhancing the model's transferability towards the continually changing target domain. We introduce a multi-task setting for PCoTTA, which is practical and realistic, handling multiple tasks within one unified model during the continual adaptation. Our PCoTTA involves three key components: automatic prototype mixture (APM), Gaussian Splatted feature shifting (GSFS), and contrastive prototype repulsion (CPR). Firstly, APM is designed to automatically mix the source prototypes with the learnable prototypes with a similarity balancing factor, avoiding catastrophic forgetting. Then, GSFS dynamically shifts the testing sample toward the source domain, mitigating error accumulation in an online manner. In addition, CPR is proposed to pull the nearest learnable prototype close to the testing feature and push it away from other prototypes, making each prototype distinguishable during the adaptation. Experimental comparisons lead to a new benchmark, demonstrating PCoTTA's superiority in boosting the model's transferability towards the continually changing target domain.