BoostPoint: Boosting point cloud backbones with image pre-training for 3D understanding

Nowadays, pre-training models on large-scale datasets and fine-tuning models on task-specific datasets have become common paradigms, achieving impressive success in natural language processing and 2D vision. Nonetheless, the potential of this paradigm has not been fully explored in 3D vision due to the scale of the datasets. To overcome this, we propose BoostPoint, a novel pipeline that uses large-scale rendered images as 3D point cloud model inputs for pre-training and uses general 3D tasks for fine-tuning. In BoostPoint, we propose a novel learning-free image-to-point (I2P) module to transform raw pixels into required inputs. Specifically, we view pixels as unorganized points, including essential raw features (e.g., color) and positional information (e.g., coordinates). Employing simple linear iterative clustering (SLIC), the I2P module effectively groups these unorganized points into superpixels, facilitating point cloud backbone pre-training. Furthermore, we employ a modality-agnostic debiasing mechanism during pre-training to prevent negative transfer in downstream tasks. Extensive fine-tuning experiments show that BoostPoint provides significant improvements to 3D point cloud backbones for 3D point cloud classification and part segmentation.