Less is More: Towards Efficient Few-shot 3D Semantic Segmentation via Training-free Networks
This work addresses efficiency and domain adaptation issues in 3D segmentation for applications like robotics and autonomous driving, offering a novel training-free approach with incremental improvements.
The paper tackles the problem of excessive time overhead and domain gaps in few-shot 3D semantic segmentation by proposing TFS3D, a training-free network, and TFS3D-T, a variant with lightweight training. TFS3D achieves comparable performance to training-based methods without learnable parameters, and TFS3D-T improves state-of-the-art by +6.93% and +17.96% mIoU on S3DIS and ScanNet while reducing training time by -90%.
To reduce the reliance on large-scale datasets, recent works in 3D segmentation resort to few-shot learning. Current 3D few-shot semantic segmentation methods first pre-train the models on `seen' classes, and then evaluate their generalization performance on `unseen' classes. However, the prior pre-training stage not only introduces excessive time overhead, but also incurs a significant domain gap on `unseen' classes. To tackle these issues, we propose an efficient Training-free Few-shot 3D Segmentation netwrok, TFS3D, and a further training-based variant, TFS3D-T. Without any learnable parameters, TFS3D extracts dense representations by trigonometric positional encodings, and achieves comparable performance to previous training-based methods. Due to the elimination of pre-training, TFS3D can alleviate the domain gap issue and save a substantial amount of time. Building upon TFS3D, TFS3D-T only requires to train a lightweight query-support transferring attention (QUEST), which enhances the interaction between the few-shot query and support data. Experiments demonstrate TFS3D-T improves previous state-of-the-art methods by +6.93% and +17.96% mIoU respectively on S3DIS and ScanNet, while reducing the training time by -90%, indicating superior effectiveness and efficiency.