动态室内环境中的少样本增量3D目标检测
Few-Shot Incremental 3D Object Detection in Dynamic Indoor Environments
把VLM的2D语义特征与类无关3D框结合成类别原型，并用多模态门控打分，为少样本增量3D检测提供了一条不依赖大量新类标注的路线；其原型印记与特征重加权机制可迁移到其他开放世界/增量检测设定。
原文依据 FI3Det introduces a VLM-guided unknown object learning module in the base stage to enhance perception of unseen categories.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论增量式3D物体感知是动态室内环境中实现具身智能的关键一步。然而，现有的增量3D检测方法依赖对新类别的广泛标注才能获得令人满意的性能。为解决这一局限，我们提出FI3Det，一个少样本增量3D检测框架，通过利用视觉语言模型（VLM）学习未见类别的知识，仅需少量新样本即可实现高效的3D感知。FI3Det在基础阶段引入一个VLM引导的未知物体学习模块，以增强对未见类别的感知。具体而言，它利用VLM挖掘未知物体并提取全面的表征，包括2D语义特征和类无关的3D边界框。为缓解这些表征中的噪声，进一步设计了一种加权机制，根据空间位置以及每个框内的特征一致性对点级和框级特征的贡献进行重加权。此外，FI3Det提出一种门控多模态原型印记模块，其中类别原型由对齐的2D语义特征与3D几何特征构建，用于计算分类得分，随后通过多模态门控机制进行融合以完成新物体检测。作为少样本增量3D目标检测的首个框架，我们在ScanNet V2和SUN RGB-D两个数据集上建立了批处理与序列两种评测设置，FI3Det在其中相较基线方法取得了显著且一致的提升。
英文原文摘要 1534 字符
Incremental 3D object perception is a critical step toward embodied intelligence in dynamic indoor environments. However, existing incremental 3D detection methods rely on extensive annotations of novel classes for satisfactory performance. To address this limitation, we propose FI3Det, a Few-shot Incremental 3D Detection framework that enables efficient 3D perception with only a few novel samples by leveraging vision-language models (VLMs) to learn knowledge of unseen categories. FI3Det introduces a VLM-guided unknown object learning module in the base stage to enhance perception of unseen categories. Specifically, it employs VLMs to mine unknown objects and extract comprehensive representations, including 2D semantic features and class-agnostic 3D bounding boxes. To mitigate noise in these representations, a weighting mechanism is further designed to re-weight the contributions of point- and box-level features based on their spatial locations and feature consistency within each box. Moreover, FI3Det proposes a gated multimodal prototype imprinting module, where category prototypes are constructed from aligned 2D semantic and 3D geometric features to compute classification scores, which are then fused via a multimodal gating mechanism for novel object detection. As the first framework for few-shot incremental 3D object detection, we establish both batch and sequential evaluation settings on two datasets, ScanNet V2 and SUN RGB-D, where FI3Det achieves strong and consistent improvements over baseline methods.