Fast3D: Accelerating 3D Multi-modal Large Language Models for Efficient 3D Scene Understanding

• URL: http://arxiv.org/abs/2507.09334v1
• Date: Sat, 12 Jul 2025 16:29:02 GMT
• Title: Fast3D: Accelerating 3D Multi-modal Large Language Models for Efficient 3D Scene Understanding
• Authors: Wencan Huang, Daizong Liu, Wei Hu,
• Abstract summary: We propose Fast3D, a plug-and-play visual token pruning framework for 3D MLLMs.<n>Global Attention Prediction (GAP) learns to predict the global attention distributions of the target model, enabling efficient token importance estimation.<n>SAP, introduces dynamic token budgets through attention-based complexity assessment, automatically adjusting layer-wise pruning ratios.
• Score: 24.964149224068027
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While 3D Multi-modal Large Language Models (MLLMs) demonstrate remarkable scene understanding capabilities, their practical deployment faces critical challenges due to computational inefficiency. The key bottleneck stems from processing excessive object-centric visual tokens required for comprehensive 3D scene representation. Although visual token pruning has shown promise in accelerating 2D MLLMs, its applicability to 3D domains remains largely unexplored due to fundamental disparities in token structures. In this paper, we reveal two critical insights: (1) Significant redundancy exists in object-level 3D token representations, analogous to patch-level redundancy in 2D systems; (2) Global attention patterns exhibit strong predictive power for identifying non-essential tokens in 3D contexts. Building on these observations, we propose Fast3D, a plug-and-play visual token pruning framework for 3D MLLMs featuring two technical innovations: (1) Global Attention Prediction (GAP), where a lightweight neural network learns to predict the global attention distributions of the target model, enabling efficient token importance estimation for precise pruning guidance; (2) Sample-Adaptive visual token Pruning (SAP), which introduces dynamic token budgets through attention-based complexity assessment, automatically adjusting layer-wise pruning ratios based on input characteristics. Both of these two techniques operate without modifying the parameters of the target model. Extensive evaluations across five benchmarks validate the effectiveness of Fast3D, particularly under high visual token pruning ratios. Code is available at https://github.com/wencan25/Fast3D

Related papers

• Pts3D-LLM: Studying the Impact of Token Structure for 3D Scene Understanding With Large Language Models [9.658828841170472]
  This work presents a rigorous study of 3D token structures, systematically comparing video-based and point-based representations.<n>We propose a novel approach that enriches visual tokens by incorporating 3D point cloud features from a Sonata pretrained Point Transformer V3 encoder.
  arXiv  Detail & Related papers  (2025-06-06T02:35:26Z)
• AdaToken-3D: Dynamic Spatial Gating for Efficient 3D Large Multimodal-Models Reasoning [27.40106634796608]
  Large Multimodal Models (LMMs) have become a pivotal research focus in deep learning.<n>Currently, 3D LMMs employing thousands of spatial tokens for multimodal reasoning suffer from critical inefficiencies.<n>We propose AdaToken-3D, an adaptive spatial token optimization framework that dynamically prunes redundant tokens.
  arXiv  Detail & Related papers  (2025-05-19T07:11:07Z)
• 3D-AffordanceLLM: Harnessing Large Language Models for Open-Vocabulary Affordance Detection in 3D Worlds [81.14476072159049]
  3D Affordance detection is a challenging problem with broad applications on various robotic tasks.<n>We reformulate the traditional affordance detection paradigm into textit Reasoning Affordance (IRAS) task.<n>We propose 3D-ADLLM, a framework designed for reasoning affordance detection in 3D open-scene.
  arXiv  Detail & Related papers  (2025-02-27T12:29:44Z)
• Articulate3D: Holistic Understanding of 3D Scenes as Universal Scene Description [56.69740649781989]
  3D scene understanding is a long-standing challenge in computer vision and a key component in enabling mixed reality, wearable computing, and embodied AI.<n>We introduce Articulate3D, an expertly curated 3D dataset featuring high-quality manual annotations on 280 indoor scenes.<n>We also present USDNet, a novel unified framework capable of simultaneously predicting part segmentation along with a full specification of motion attributes for articulated objects.
  arXiv  Detail & Related papers  (2024-12-02T11:33:55Z)
• MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations [55.022519020409405]
  This paper builds the first largest ever multi-modal 3D scene dataset and benchmark with hierarchical grounded language annotations, MMScan.<n>The resulting multi-modal 3D dataset encompasses 1.4M meta-annotated captions on 109k objects and 7.7k regions as well as over 3.04M diverse samples for 3D visual grounding and question-answering benchmarks.
  arXiv  Detail & Related papers  (2024-06-13T17:59:30Z)
• Is a 3D-Tokenized LLM the Key to Reliable Autonomous Driving? [66.6886931183372]
  We introduce DETR-style 3D perceptrons as 3D tokenizers, which connect LLM with a one-layer linear projector. Despite its simplicity, Atlas demonstrates superior performance in both 3D detection and ego planning tasks.
  arXiv  Detail & Related papers  (2024-05-28T16:57:44Z)
• Generalized Robot 3D Vision-Language Model with Fast Rendering and Pre-Training Vision-Language Alignment [55.11291053011696]
  This work presents a framework for dealing with 3D scene understanding when the labeled scenes are quite limited.<n>To extract knowledge for novel categories from the pre-trained vision-language models, we propose a hierarchical feature-aligned pre-training and knowledge distillation strategy.<n>In the limited reconstruction case, our proposed approach, termed WS3D++, ranks 1st on the large-scale ScanNet benchmark.
  arXiv  Detail & Related papers  (2023-12-01T15:47:04Z)
• LocATe: End-to-end Localization of Actions in 3D with Transformers [91.28982770522329]
  LocATe is an end-to-end approach that jointly localizes and recognizes actions in a 3D sequence. Unlike transformer-based object-detection and classification models which consider image or patch features as input, LocATe's transformer model is capable of capturing long-term correlations between actions in a sequence. We introduce a new, challenging, and more realistic benchmark dataset, BABEL-TAL-20 (BT20), where the performance of state-of-the-art methods is significantly worse.
  arXiv  Detail & Related papers  (2022-03-21T03:35:32Z)
• SA-Det3D: Self-Attention Based Context-Aware 3D Object Detection [9.924083358178239]
  We propose two variants of self-attention for contextual modeling in 3D object detection. We first incorporate the pairwise self-attention mechanism into the current state-of-the-art BEV, voxel and point-based detectors. Next, we propose a self-attention variant that samples a subset of the most representative features by learning deformations over randomly sampled locations.
  arXiv  Detail & Related papers  (2021-01-07T18:30:32Z)
• Pointwise Attention-Based Atrous Convolutional Neural Networks [15.499267533387039]
  A pointwise attention-based atrous convolutional neural network architecture is proposed to efficiently deal with a large number of points. The proposed model has been evaluated on the two most important 3D point cloud datasets for the 3D semantic segmentation task. It achieves a reasonable performance compared to state-of-the-art models in terms of accuracy, with a much smaller number of parameters.
  arXiv  Detail & Related papers  (2019-12-27T13:12:58Z)

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.