Seeing 3D Through 2D Lenses: 3D Few-Shot Class-Incremental Learning via Cross-Modal Geometric Rectification
This work addresses the need for expandable 3D recognition systems in open-world scenarios, offering a novel solution to mitigate catastrophic forgetting and semantic blurring in incremental learning, though it is domain-specific to 3D vision.
The paper tackled the problem of 3D few-shot class-incremental learning, which struggles with geometric misalignment and texture bias under data scarcity, by proposing the Cross-Modal Geometric Rectification (CMGR) framework, resulting in significant improvements in geometric coherence and robustness across cross-domain and within-domain settings.
The rapid growth of 3D digital content necessitates expandable recognition systems for open-world scenarios. However, existing 3D class-incremental learning methods struggle under extreme data scarcity due to geometric misalignment and texture bias. While recent approaches integrate 3D data with 2D foundation models (e.g., CLIP), they suffer from semantic blurring caused by texture-biased projections and indiscriminate fusion of geometric-textural cues, leading to unstable decision prototypes and catastrophic forgetting. To address these issues, we propose Cross-Modal Geometric Rectification (CMGR), a framework that enhances 3D geometric fidelity by leveraging CLIP's hierarchical spatial semantics. Specifically, we introduce a Structure-Aware Geometric Rectification module that hierarchically aligns 3D part structures with CLIP's intermediate spatial priors through attention-driven geometric fusion. Additionally, a Texture Amplification Module synthesizes minimal yet discriminative textures to suppress noise and reinforce cross-modal consistency. To further stabilize incremental prototypes, we employ a Base-Novel Discriminator that isolates geometric variations. Extensive experiments demonstrate that our method significantly improves 3D few-shot class-incremental learning, achieving superior geometric coherence and robustness to texture bias across cross-domain and within-domain settings.