自适应查询与位置编码，让3D实例分割模型更好应对不同大小场景。
AQ3D: Adaptive Query Transformer for 3D Instance Segmentation
- 按场景超点比例动态生成查询，避免固定数量带来的过/欠初始化问题。
- 用量化坐标上的3D RoPE替代学习的定位表，提升位置表示能力。
- 引入属性池化、掩码优化分支和余弦分类器，显著提升分割精度。
基于Transformer的3D实例分割解码器通常采用固定数量的查询和在训练分布上校准的位置编码，而室内扫描在空间范围和物体数量上差异极大，导致固定查询集在小场景中过初始化、大场景中欠初始化，且学习的绝对/相对编码受限于训练场景范围，易饱和。本文提出AQ3D，可在训练和推理时适应不同大小的场景。查询以场景超点数量的固定比例生成，形成过完备集合，背景剔除完全由解码器完成。位置信息通过量化度量坐标的3D RoPE编码，取代先前解码器中学习的有界查找表。此外，通过属性驱动的超点池化、掩码精修分支和余弦分类器改进解码器结构。实验表明，本方法在未使用额外数据增强的前提下，在ScanNetV2、ScanNet200和ScanNet++V2数据集的验证集与隐藏测试集上达到新最优性能。代码已开源。
原文摘要 · Abstract (English)
Transformer-based decoders for 3D instance segmentation typically commit to a fixed number of queries and positional modeling calibrated on the training distribution rather than on the scene at hand. Indoor scans vary widely in spatial extent and object count, so a fixed query set over-initializes small scenes and under-initializes large ones, while learned absolute and relative encodings are bound to the training scenes' extents and can saturate. We present AQ3D, which is designed to handle scenes of various sizes during training and inference. Queries are instantiated at a fixed ratio of the scene's superpoints, forming an overcomplete set whose background rejection is entirely left to the decoder. Positional information is encoded using 3D RoPE over quantized metric coordinates, replacing learned bounded lookup tables of prior decoders. Further, we improve the decoder itself by using attribution-based superpoint pooling, a mask refinement branch, and a cosine classifier for background rejection. Experiments show our method sets a new state-of-the-art on validation and hidden test splits across the datasets ScanNetV2, ScanNet200, and ScanNet++V2 among decoder methods trained without additional data augmentation. Code is available at \href{https://github.com/kenomo/aq3d}{github.com/kenomo/aq3d}.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。