PMA: Towards Parameter-Efficient Point Cloud Understanding via Point Mamba Adapter
This work addresses a bottleneck in 3D perception for researchers and practitioners by enhancing parameter efficiency and feature integration, though it is incremental as it builds on existing pre-trained model paradigms.
The paper tackles the problem of underutilizing intermediate layer information in pre-trained models for point cloud understanding by proposing Point Mamba Adapter (PMA), which fuses multi-layer features using Mamba and a geometry-constrained gate prompt generator, achieving improved performance on challenging datasets.
Applying pre-trained models to assist point cloud understanding has recently become a mainstream paradigm in 3D perception. However, existing application strategies are straightforward, utilizing only the final output of the pre-trained model for various task heads. It neglects the rich complementary information in the intermediate layer, thereby failing to fully unlock the potential of pre-trained models. To overcome this limitation, we propose an orthogonal solution: Point Mamba Adapter (PMA), which constructs an ordered feature sequence from all layers of the pre-trained model and leverages Mamba to fuse all complementary semantics, thereby promoting comprehensive point cloud understanding. Constructing this ordered sequence is non-trivial due to the inherent isotropy of 3D space. Therefore, we further propose a geometry-constrained gate prompt generator (G2PG) shared across different layers, which applies shared geometric constraints to the output gates of the Mamba and dynamically optimizes the spatial order, thus enabling more effective integration of multi-layer information. Extensive experiments conducted on challenging point cloud datasets across various tasks demonstrate that our PMA elevates the capability for point cloud understanding to a new level by fusing diverse complementary intermediate features. Code is available at https://github.com/zyh16143998882/PMA.