EDA: Explicit Text-Decoupling and Dense Alignment for 3D Visual Grounding
This work improves 3D visual grounding for applications like robotics and augmented reality by introducing a method that decouples textual attributes for better alignment, though it is incremental in nature.
The paper tackles the problem of 3D visual grounding by addressing the loss of word-level information and neglect of attributes in existing methods, achieving state-of-the-art performance on datasets like ScanRefer and SR3D/NR3D.
3D visual grounding aims to find the object within point clouds mentioned by free-form natural language descriptions with rich semantic cues. However, existing methods either extract the sentence-level features coupling all words or focus more on object names, which would lose the word-level information or neglect other attributes. To alleviate these issues, we present EDA that Explicitly Decouples the textual attributes in a sentence and conducts Dense Alignment between such fine-grained language and point cloud objects. Specifically, we first propose a text decoupling module to produce textual features for every semantic component. Then, we design two losses to supervise the dense matching between two modalities: position alignment loss and semantic alignment loss. On top of that, we further introduce a new visual grounding task, locating objects without object names, which can thoroughly evaluate the model's dense alignment capacity. Through experiments, we achieve state-of-the-art performance on two widely-adopted 3D visual grounding datasets, ScanRefer and SR3D/NR3D, and obtain absolute leadership on our newly-proposed task. The source code is available at https://github.com/yanmin-wu/EDA.