Joint prototype and coefficient prediction for 3D instance segmentation
This addresses practical 3D scene understanding applications requiring rapid and reliable inference, representing a strong domain-specific advancement.
The paper tackles 3D instance segmentation by introducing a method that simultaneously learns coefficients and prototypes with an overcomplete sampling strategy, achieving superior performance on S3DIS-blocks with 32.9% faster inference and over 20-fold reduction in inference time variance compared to state-of-the-art methods.
3D instance segmentation is crucial for applications demanding comprehensive 3D scene understanding. In this paper, we introduce a novel method that simultaneously learns coefficients and prototypes. Employing an overcomplete sampling strategy, our method produces an overcomplete set of instance predictions, from which the optimal ones are selected through a Non-Maximum Suppression (NMS) algorithm during inference. The obtained prototypes are visualizable and interpretable. Our method demonstrates superior performance on S3DIS-blocks, consistently outperforming existing methods in mRec and mPrec. Moreover, it operates 32.9% faster than the state-of-the-art. Notably, with only 0.8% of the total inference time, our method exhibits an over 20-fold reduction in the variance of inference time compared to existing methods. These attributes render our method well-suited for practical applications requiring both rapid inference and high reliability.