LLMI3D: MLLM-based 3D Perception from a Single 2D Image
- URL: http://arxiv.org/abs/2408.07422v2
- Date: Thu, 13 Feb 2025 09:32:44 GMT
- Title: LLMI3D: MLLM-based 3D Perception from a Single 2D Image
- Authors: Fan Yang, Sicheng Zhao, Yanhao Zhang, Hui Chen, Haonan Lu, Jungong Han, Guiguang Ding,
- Abstract summary: multimodal large language models (MLLMs) excel in general capacity but underperform in 3D tasks.
In this paper, we propose solutions for weak 3D local spatial object perception, poor text-based geometric numerical output, and inability to handle camera focal variations.
We employ parameter-efficient fine-tuning for a pre-trained MLLM and develop LLMI3D, a powerful 3D perception MLLM.
- Score: 77.13869413871028
- License:
- Abstract: Recent advancements in autonomous driving, augmented reality, robotics, and embodied intelligence have necessitated 3D perception algorithms. However, current 3D perception methods, especially specialized small models, exhibit poor generalization in open scenarios. On the other hand, multimodal large language models (MLLMs) excel in general capacity but underperform in 3D tasks, due to weak 3D local spatial object perception, poor text-based geometric numerical output, and inability to handle camera focal variations. To address these challenges, we propose the following solutions: Spatial-Enhanced Local Feature Mining for better spatial feature extraction, 3D Query Token-Derived Info Decoding for precise geometric regression, and Geometry Projection-Based 3D Reasoning for handling camera focal length variations. We employ parameter-efficient fine-tuning for a pre-trained MLLM and develop LLMI3D, a powerful 3D perception MLLM. Additionally, we have constructed the IG3D dataset, which provides fine-grained descriptions and question-answer annotations. Extensive experiments demonstrate that our LLMI3D achieves state-of-the-art performance, outperforming other methods by a large margin.
Related papers
- 3D-MoE: A Mixture-of-Experts Multi-modal LLM for 3D Vision and Pose Diffusion via Rectified Flow [69.94527569577295]
3D vision and spatial reasoning have long been recognized as preferable for accurately perceiving our three-dimensional world.
Due to the difficulties in collecting high-quality 3D data, research in this area has only recently gained momentum.
We propose converting existing densely activated LLMs into mixture-of-experts (MoE) models, which have proven effective for multi-modal data processing.
arXiv Detail & Related papers (2025-01-28T04:31:19Z) - 3UR-LLM: An End-to-End Multimodal Large Language Model for 3D Scene Understanding [49.15555885075644]
We develop pipeline based on open-source 2D MLLMs and LLMs to generate high-quality 3D-text pairs.
We introduce the 3UR-LLM model, an end-to-end 3D MLLM designed for precise interpretation of 3D scenes.
arXiv Detail & Related papers (2025-01-14T03:50:23Z) - 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.
Our model achieves state-of-the-art performance on several 3D scene understanding benchmarks.
arXiv Detail & Related papers (2024-11-30T14:28:53Z) - Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation [30.744137117668643]
Lift3D is a framework that enhances 2D foundation models with implicit and explicit 3D robotic representations to construct a robust 3D manipulation policy.
In experiments, Lift3D consistently outperforms previous state-of-the-art methods across several simulation benchmarks and real-world scenarios.
arXiv Detail & Related papers (2024-11-27T18:59:52Z) - BIP3D: Bridging 2D Images and 3D Perception for Embodied Intelligence [11.91274849875519]
We introduce a novel image-centric 3D perception model, BIP3D, to overcome the limitations of point-centric methods.
We leverage pre-trained 2D vision foundation models to enhance semantic understanding, and introduce a spatial enhancer module to improve spatial understanding.
In our experiments, BIP3D outperforms current state-of-the-art results on the EmbodiedScan benchmark, achieving improvements of 5.69% in the 3D detection task and 15.25% in the 3D visual grounding task.
arXiv Detail & Related papers (2024-11-22T11:35:42Z) - SPARTUN3D: Situated Spatial Understanding of 3D World in Large Language Models [45.28780381341979]
We introduce a scalable situated 3D dataset, named Spartun3D, that incorporates various situated spatial reasoning tasks.
We also propose Spartun3D-LLM, built on an existing 3D-based LLM but integrated with a novel situated spatial alignment module.
arXiv Detail & Related papers (2024-10-04T19:22:20Z) - MM3DGS SLAM: Multi-modal 3D Gaussian Splatting for SLAM Using Vision, Depth, and Inertial Measurements [59.70107451308687]
We show for the first time that using 3D Gaussians for map representation with unposed camera images and inertial measurements can enable accurate SLAM.
Our method, MM3DGS, addresses the limitations of prior rendering by enabling faster scale awareness, and improved trajectory tracking.
We also release a multi-modal dataset, UT-MM, collected from a mobile robot equipped with a camera and an inertial measurement unit.
arXiv Detail & Related papers (2024-04-01T04:57:41Z) - AutoDecoding Latent 3D Diffusion Models [95.7279510847827]
We present a novel approach to the generation of static and articulated 3D assets that has a 3D autodecoder at its core.
The 3D autodecoder framework embeds properties learned from the target dataset in the latent space.
We then identify the appropriate intermediate volumetric latent space, and introduce robust normalization and de-normalization operations.
arXiv Detail & Related papers (2023-07-07T17:59:14Z) - Lightweight Multi-View 3D Pose Estimation through Camera-Disentangled
Representation [57.11299763566534]
We present a solution to recover 3D pose from multi-view images captured with spatially calibrated cameras.
We exploit 3D geometry to fuse input images into a unified latent representation of pose, which is disentangled from camera view-points.
Our architecture then conditions the learned representation on camera projection operators to produce accurate per-view 2d detections.
arXiv Detail & Related papers (2020-04-05T12:52:29Z)
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.