Does Your 3D Encoder Really Work? When Pretrain-SFT from 2D VLMs Meets 3D VLMs

• URL: http://arxiv.org/abs/2506.05318v2
• Date: Fri, 06 Jun 2025 07:09:06 GMT
• Title: Does Your 3D Encoder Really Work? When Pretrain-SFT from 2D VLMs Meets 3D VLMs
• Authors: Haoyuan Li, Yanpeng Zhou, Yufei Gao, Tao Tang, Jianhua Han, Yujie Yuan, Dave Zhenyu Chen, Jiawang Bian, Hang Xu, Xiaodan Liang,
• Abstract summary: This paper categorizes recent 3D Vision-Language Models into 3D object-centric, 2D image-based, and 3D scene-centric approaches.<n>Despite the architectural similarity of 3D scene-centric VLMs to their 2D counterparts, they have exhibited comparatively lower performance compared with the latest 3D object-centric and 2D image-based approaches.<n>Our investigation suggests that while these models possess cross-modal alignment capabilities, they tend to over-rely on linguistic cues and overfit to frequent answer distributions.
• Score: 72.11701578308804
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Remarkable progress in 2D Vision-Language Models (VLMs) has spurred interest in extending them to 3D settings for tasks like 3D Question Answering, Dense Captioning, and Visual Grounding. Unlike 2D VLMs that typically process images through an image encoder, 3D scenes, with their intricate spatial structures, allow for diverse model architectures. Based on their encoder design, this paper categorizes recent 3D VLMs into 3D object-centric, 2D image-based, and 3D scene-centric approaches. Despite the architectural similarity of 3D scene-centric VLMs to their 2D counterparts, they have exhibited comparatively lower performance compared with the latest 3D object-centric and 2D image-based approaches. To understand this gap, we conduct an in-depth analysis, revealing that 3D scene-centric VLMs show limited reliance on the 3D scene encoder, and the pre-train stage appears less effective than in 2D VLMs. Furthermore, we observe that data scaling benefits are less pronounced on larger datasets. Our investigation suggests that while these models possess cross-modal alignment capabilities, they tend to over-rely on linguistic cues and overfit to frequent answer distributions, thereby diminishing the effective utilization of the 3D encoder. To address these limitations and encourage genuine 3D scene understanding, we introduce a novel 3D Relevance Discrimination QA dataset designed to disrupt shortcut learning and improve 3D understanding. Our findings highlight the need for advanced evaluation and improved strategies for better 3D understanding in 3D VLMs.

Related papers

• Unifying 2D and 3D Vision-Language Understanding [85.84054120018625]
  We introduce UniVLG, a unified architecture for 2D and 3D vision-language learning.<n>UniVLG bridges the gap between existing 2D-centric models and the rich 3D sensory data available in embodied systems.
  arXiv  Detail & Related papers  (2025-03-13T17:56:22Z)
• SplatTalk: 3D VQA with Gaussian Splatting [13.211810095081159]
  Language-guided 3D scene understanding is important for advancing applications in robotics, AR/VR, and human-computer interaction.<n>We introduce SplatTalk, a novel method that uses a generalizable 3D Gaussian Splatting (3DGS) framework to produce 3D tokens suitable for direct input into a pretrained LLM.
  arXiv  Detail & Related papers  (2025-03-08T16:31:48Z)
• Inst3D-LMM: Instance-Aware 3D Scene Understanding with Multi-modal Instruction Tuning [18.185457833299235]
  We propose a unified Instance-aware 3D Large Multi-modal Model (Inst3D-LMM) to deal with multiple 3D scene understanding tasks simultaneously.<n>We first introduce a novel Multi-view Cross-Modal Fusion (MCMF) module to inject the multi-view 2D semantics into their corresponding 3D geometric features.<n>For scene-level relation-aware tokens, we further present a 3D Instance Spatial Relation (3D-ISR) module to capture the intricate pairwise spatial relationships among objects.
  arXiv  Detail & Related papers  (2025-03-01T14:38:42Z)
• Video-3D LLM: Learning Position-Aware Video Representation for 3D Scene Understanding [19.382210260928776]
  Video-3D LLM treats 3D scenes as dynamic videos and incorporates 3D position encoding into these representations.<n>Our model achieves state-of-the-art performance on several 3D scene understanding benchmarks.
  arXiv  Detail & Related papers  (2024-11-30T14:28:53Z)
• LLaVA-3D: A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness [22.408933972095763]
  We introduce a simple yet effective framework called LLaVA-3D.<n>It efficiently adapts LLaVA for 3D scene understanding without compromising 2D understanding capabilities.<n>LLaVA-3D converges 3.5x faster than existing 3D LMMs when trained on 3D vision-language datasets.
  arXiv  Detail & Related papers  (2024-09-26T17:59:11Z)
• Uni3D: Exploring Unified 3D Representation at Scale [66.26710717073372]
  We present Uni3D, a 3D foundation model to explore the unified 3D representation at scale. Uni3D uses a 2D ViT end-to-end pretrained to align the 3D point cloud features with the image-text aligned features. We show that the strong Uni3D representation also enables applications such as 3D painting and retrieval in the wild.
  arXiv  Detail & Related papers  (2023-10-10T16:49:21Z)
• Chat-3D: Data-efficiently Tuning Large Language Model for Universal Dialogue of 3D Scenes [56.727745047799246]
  3D scene understanding has gained significant attention due to its wide range of applications. This paper presents Chat-3D, which combines the 3D visual perceptual ability of pre-trained 3D representations and the impressive reasoning and conversation capabilities of advanced LLMs.
  arXiv  Detail & Related papers  (2023-08-17T03:52:15Z)
• 3D-LLM: Injecting the 3D World into Large Language Models [60.43823088804661]
  Large language models (LLMs) and Vision-Language Models (VLMs) have been proven to excel at multiple tasks, such as commonsense reasoning. We propose to inject the 3D world into large language models and introduce a new family of 3D-LLMs. Specifically, 3D-LLMs can take 3D point clouds and their features as input and perform a diverse set of 3D-related tasks.
  arXiv  Detail & Related papers  (2023-07-24T17:59:02Z)
• 3D-to-2D Distillation for Indoor Scene Parsing [78.36781565047656]
  We present a new approach that enables us to leverage 3D features extracted from large-scale 3D data repository to enhance 2D features extracted from RGB images. First, we distill 3D knowledge from a pretrained 3D network to supervise a 2D network to learn simulated 3D features from 2D features during the training. Second, we design a two-stage dimension normalization scheme to calibrate the 2D and 3D features for better integration. Third, we design a semantic-aware adversarial training model to extend our framework for training with unpaired 3D data.
  arXiv  Detail & Related papers  (2021-04-06T02:22:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.