Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Model

• URL: http://arxiv.org/abs/2605.14950v1
• Date: Thu, 14 May 2026 15:21:36 GMT
• Title: Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Model
• Abstract summary: Vision-Language-Action models are promising for unifying perception, language grounding, and action generation.<n>Current VLA models rely primarily on 2D visual representations that lack depth information and detailed spatial relationships.<n>Evo-Depth is a lightweight depth-enhanced VLA framework that enhances spatially grounded manipulation without relying on additional sensing or compromising deployment efficiency.
• Score: 43.14057937517956
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language-Action models have emerged as a promising paradigm for robotic manipulation by unifying perception, language grounding, and action generation. However, they often struggle in scenarios requiring precise spatial understanding, as current VLA models primarily rely on 2D visual representations that lack depth information and detailed spatial relationships. While recent approaches incorporate explicit 3D inputs such as depth maps or point clouds to address this issue, they often increase system complexity, require additional sensors, and remain vulnerable to sensing noise and reconstruction errors. Another line of work explores implicit 3D-aware spatial modeling directly from RGB observations without extra sensors, but it often relies on large geometry foundation models, resulting in higher training and deployment costs. To address these challenges, we propose Evo-Depth, a lightweight depth-enhanced VLA framework that enhances spatially grounded manipulation without relying on additional sensing hardware or compromising deployment efficiency. Evo-Depth employs a lightweight Implicit Depth Encoding Module to extract compact depth features from multi-view RGB images. These features are incorporated into vision-language representations through a Spatial Enhancement Module via depth-aware modulation, enabling efficient spatial-semantic enhancement. A Progressive Alignment Training strategy is further introduced to align the resulting depth-enhanced representations with downstream action learning. With only 0.9B parameters, Evo-Depth achieves superior performance across four simulation benchmarks. In real-world experiments, Evo-Depth attains the highest average success rate while also exhibiting the smallest model size, lowest GPU memory usage, and highest inference frequency among compared methods.

Related papers

• GA-VLN: Geometry-Aware BEV Representation for Efficient Vision-Language Navigation [75.85672467847631]
  We introduce the Geometry-Aware BEV (GA-BEV) - a compact, 3D-grounded feature representation.<n>We construct BEV spatial maps from RGB-D inputs by projecting visual features into 3D space.<n>We incorporate features from a pretrained 3D foundation model into the BEV space, injecting structural priors learned from large-scale 3D reconstruction tasks.
  arXiv  Detail & Related papers  (2026-05-21T06:20:17Z)
• PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction [57.63073414949329]
  We propose PointACT, a dual-system 3D-aware VLA policy that integrates hierarchical 3D point cloud representations directly into the action decoding process.<n>PointACT employs a multi-scale point-action interaction mechanism with efficient bottleneck window self-attention, enabling evolving action tokens to densely attend to both local geometric detail and global scene structure.
  arXiv  Detail & Related papers  (2026-05-20T17:10:31Z)
• VEGA: Visual Encoder Grounding Alignment for Spatially-Aware Vision-Language-Action Models [55.12929235609365]
  Visual backbones of current vision-language-action (VLA) models are predominantly pretrained on 2D image data without explicit 3D geometric supervision.<n>Existing implicit grounding methods partially address this by aligning VLA features with those of 3D-aware foundation models.<n>We propose VEGA, a framework that directly aligns the output of the VLA's visual encoder with spatially-aware features from DINOv2-FiT3D.
  arXiv  Detail & Related papers  (2026-05-11T12:44:26Z)
• AugVLA-3D: Depth-Driven Feature Augmentation for Vision-Language-Action Models [42.57469056850227]
  Vision-Language-Action (VLA) models have recently achieved remarkable progress in robotic perception and control.<n>We propose a novel framework that integrates depth estimation into VLA models to enrich 3D feature representations.
  arXiv  Detail & Related papers  (2026-02-11T09:57:32Z)
• Abstract 3D Perception for Spatial Intelligence in Vision-Language Models [100.13033631690114]
  Vision-language models (VLMs) struggle with 3D-related tasks such as spatial cognition and physical understanding.<n>We introduce SandboxVLM, a framework that leverages abstract bounding boxes to encode geometric structure and physical kinematics for VLM.<n>Our approach consistently improves spatial intelligence, achieving an 8.3% gain on SAT Real compared with baseline methods.
  arXiv  Detail & Related papers  (2025-11-14T04:16:09Z)
• Move to Understand a 3D Scene: Bridging Visual Grounding and Exploration for Efficient and Versatile Embodied Navigation [54.04601077224252]
  Embodied scene understanding requires not only comprehending visual-spatial information but also determining where to explore next in the 3D physical world.<n>underlinetextbf3D vision-language learning enables embodied agents to effectively explore and understand their environment.<n>model's versatility enables navigation using diverse input modalities, including categories, language descriptions, and reference images.
  arXiv  Detail & Related papers  (2025-07-05T14:15:52Z)
• Evo-0: Vision-Language-Action Model with Implicit Spatial Understanding [11.222744122842023]
  We introduce a plug-and-play module that implicitly incorporates 3D geometry features into Vision-Language-Action (VLA) models.<n>Our method significantly improves the performance of state-of-the-art VLA models across diverse scenarios.
  arXiv  Detail & Related papers  (2025-07-01T04:05:47Z)
• Lightweight RGB-D Salient Object Detection from a Speed-Accuracy Tradeoff Perspective [54.91271106816616]
  Current RGB-D methods usually leverage large-scale backbones to improve accuracy but sacrifice efficiency.<n>We propose a Speed-Accuracy Tradeoff Network (SATNet) for Lightweight RGB-D SOD from three fundamental perspectives.<n> Concerning depth quality, we introduce the Depth Anything Model to generate high-quality depth maps.<n>For modality fusion, we propose a Decoupled Attention Module (DAM) to explore the consistency within and between modalities.<n>For feature representation, we develop a Dual Information Representation Module (DIRM) with a bi-directional inverted framework.
  arXiv  Detail & Related papers  (2025-05-07T19:37:20Z)
• Towards Ambiguity-Free Spatial Foundation Model: Rethinking and Decoupling Depth Ambiguity [20.86484181698326]
  Existing models, limited to deterministic predictions, overlook real-world multi-layer depth.<n>We introduce a paradigm shift from single-prediction to multi-hypothesis spatial foundation models.
  arXiv  Detail & Related papers  (2025-03-08T02:33:54Z)

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.