AdvDreamer Unveils: Are Vision-Language Models Truly Ready for Real-World 3D Variations?

• URL: http://arxiv.org/abs/2412.03002v3
• Date: Sun, 09 Mar 2025 13:26:29 GMT
• Title: AdvDreamer Unveils: Are Vision-Language Models Truly Ready for Real-World 3D Variations?
• Authors: Shouwei Ruan, Hanqing Liu, Yao Huang, Xiaoqi Wang, Caixin Kang, Hang Su, Yinpeng Dong, Xingxing Wei,
• Abstract summary: Vision Language Models (VLMs) have exhibited remarkable generalization capabilities, yet their robustness in dynamic real-world scenarios remains largely unexplored.<n>We propose AdvDreamer, the first framework capable of generating physically reproducible Adversarial 3D Transformation (Adv-3DT) samples from single-view observations.
• Score: 32.57500114041348
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision Language Models (VLMs) have exhibited remarkable generalization capabilities, yet their robustness in dynamic real-world scenarios remains largely unexplored. To systematically evaluate VLMs' robustness to real-world 3D variations, we propose AdvDreamer, the first framework capable of generating physically reproducible Adversarial 3D Transformation (Adv-3DT) samples from single-view observations. In AdvDreamer, we integrate three key innovations: Firstly, to characterize real-world 3D variations with limited prior knowledge precisely, we design a zero-shot Monocular Pose Manipulation pipeline built upon generative 3D priors. Secondly, to ensure the visual quality of worst-case Adv-3DT samples, we propose a Naturalness Reward Model that provides continuous naturalness regularization during adversarial optimization, effectively preventing convergence to hallucinated or unnatural elements. Thirdly, to enable systematic evaluation across diverse VLM architectures and visual-language tasks, we introduce the Inverse Semantic Probability loss as the adversarial optimization objective, which solely operates in the fundamental visual-textual alignment space. Based on the captured Adv-3DT samples with high aggressiveness and transferability, we establish MM3DTBench, the first VQA benchmark dataset tailored to evaluate VLM robustness under challenging 3D variations. Extensive evaluations of representative VLMs with varying architectures reveal that real-world 3D variations can pose severe threats to model performance across various tasks.

Related papers

• GATE3D: Generalized Attention-based Task-synergized Estimation in 3D* [0.7373617024876725]
  GATE3D is a novel framework for generalized monocular 3D object detection via weak supervision. Our results demonstrate that GATE3D significantly accelerates learning from limited annotated data.
  arXiv  Detail & Related papers  (2025-04-15T09:37:54Z)
• Multiview Equivariance Improves 3D Correspondence Understanding with Minimal Feature Finetuning [10.51523800391153]
  We evaluate and enhance the 3D awareness of ViT-based models. We propose a simple yet effective finetuning strategy based on 3D correspondences.
  arXiv  Detail & Related papers  (2024-11-29T04:02:11Z)
• ReVLA: Reverting Visual Domain Limitation of Robotic Foundation Models [55.07988373824348]
  We study the visual generalization capabilities of three existing robotic foundation models. Our study shows that the existing models do not exhibit robustness to visual out-of-domain scenarios. We propose a gradual backbone reversal approach founded on model merging.
  arXiv  Detail & Related papers  (2024-09-23T17:47:59Z)
• REVISION: Rendering Tools Enable Spatial Fidelity in Vision-Language Models [67.55362046790512]
  Vision-language models lack the ability to correctly reason over spatial relationships. We develop the REVISION framework which improves spatial fidelity in vision-language models. Our results and findings indicate that utilizing rendering-based frameworks is an effective approach for developing spatially-aware models.
  arXiv  Detail & Related papers  (2024-08-05T04:51:46Z)
• Enhancing Generalizability of Representation Learning for Data-Efficient 3D Scene Understanding [50.448520056844885]
  We propose a generative Bayesian network to produce diverse synthetic scenes with real-world patterns. A series of experiments robustly display our method's consistent superiority over existing state-of-the-art pre-training approaches.
  arXiv  Detail & Related papers  (2024-06-17T07:43:53Z)
• UPose3D: Uncertainty-Aware 3D Human Pose Estimation with Cross-View and Temporal Cues [55.69339788566899]
  UPose3D is a novel approach for multi-view 3D human pose estimation. It improves robustness and flexibility without requiring direct 3D annotations.
  arXiv  Detail & Related papers  (2024-04-23T00:18:00Z)
• Volumetric Environment Representation for Vision-Language Navigation [66.04379819772764]
  Vision-language navigation (VLN) requires an agent to navigate through a 3D environment based on visual observations and natural language instructions. We introduce a Volumetric Environment Representation (VER), which voxelizes the physical world into structured 3D cells. VER predicts 3D occupancy, 3D room layout, and 3D bounding boxes jointly.
  arXiv  Detail & Related papers  (2024-03-21T06:14:46Z)
• 3D-VLA: A 3D Vision-Language-Action Generative World Model [68.0388311799959]
  Recent vision-language-action (VLA) models rely on 2D inputs, lacking integration with the broader realm of the 3D physical world. We propose 3D-VLA by introducing a new family of embodied foundation models that seamlessly link 3D perception, reasoning, and action. Our experiments on held-in datasets demonstrate that 3D-VLA significantly improves the reasoning, multimodal generation, and planning capabilities in embodied environments.
  arXiv  Detail & Related papers  (2024-03-14T17:58:41Z)
• Retrieval-Augmented Score Distillation for Text-to-3D Generation [30.57225047257049]
  We introduce novel framework for retrieval-based quality enhancement in text-to-3D generation. We conduct extensive experiments to demonstrate that ReDream exhibits superior quality with increased geometric consistency.
  arXiv  Detail & Related papers  (2024-02-05T12:50:30Z)
• Towards Transferable Targeted 3D Adversarial Attack in the Physical World [34.36328985344749]
  transferable targeted adversarial attacks could pose a greater threat to security-critical tasks. We develop a novel framework named TT3D that could rapidly reconstruct from few multi-view images into Transferable Targeted 3D textured meshes. Experimental results show that TT3D not only exhibits superior cross-model transferability but also maintains considerable adaptability across different renders and vision tasks.
  arXiv  Detail & Related papers  (2023-12-15T06:33:14Z)
• Leveraging Large-Scale Pretrained Vision Foundation Models for Label-Efficient 3D Point Cloud Segmentation [67.07112533415116]
  We present a novel framework that adapts various foundational models for the 3D point cloud segmentation task. Our approach involves making initial predictions of 2D semantic masks using different large vision models. To generate robust 3D semantic pseudo labels, we introduce a semantic label fusion strategy that effectively combines all the results via voting.
  arXiv  Detail & Related papers  (2023-11-03T15:41:15Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [114.47216525866435]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• Progressive Multi-view Human Mesh Recovery with Self-Supervision [68.60019434498703]
  Existing solutions typically suffer from poor generalization performance to new settings. We propose a novel simulation-based training pipeline for multi-view human mesh recovery.
  arXiv  Detail & Related papers  (2022-12-10T06:28:29Z)

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.