From Scene to Object: Text-Guided Dual-Gaze Prediction

• URL: http://arxiv.org/abs/2604.20191v2
• Date: Tue, 28 Apr 2026 03:54:42 GMT
• Title: From Scene to Object: Text-Guided Dual-Gaze Prediction
• Abstract summary: Interpretable driver attention prediction is crucial for human-like autonomous driving.<n>Existing datasets provide only scene-level global gaze rather than fine-grained object-level annotations.<n>This paper proposes a novel dual-branch gaze prediction framework, establishing a complete paradigm from data construction to model architecture.
• Score: 17.32439183328327
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Interpretable driver attention prediction is crucial for human-like autonomous driving. However, existing datasets provide only scene-level global gaze rather than fine-grained object-level annotations, inherently failing to support text-grounded cognitive modeling. Consequently, while Vision-Language Models (VLMs) hold great potential for semantic reasoning, this critical data limitations leads to severe text-vision decoupling and visual-bias hallucinations. To break this bottleneck and achieve precise object-level attention prediction, this paper proposes a novel dual-branch gaze prediction framework, establishing a complete paradigm from data construction to model architecture. First, we construct G-W3DA, a object-level driver attention dataset. By integrating a multimodal large language model with the Segment Anything Model 3 (SAM3), we decouple macroscopic heatmaps into object-level masks under rigorous cross-validation, fundamentally eliminating annotation hallucinations. Building upon this high-quality data foundation, we propose the DualGaze-VLM architecture. This architecture extracts the hidden states of semantic queries and dynamically modulates visual features via a Condition-Aware SE-Gate, achieving intent-driven precise spatial anchoring. Extensive experiments on the W3DA benchmark demonstrate that DualGaze-VLM consistently surpasses existing state-of-the-art (SOTA) models in spatial alignment metrics, notably achieving up to a 17.8% improvement in Similarity (SIM) under safety-critical scenarios. Furthermore, a visual Turing test reveals that the attention heatmaps generated by DualGaze-VLM are perceived as authentic by 88.22% of human evaluators, proving its capability to generate rational cognitive priors.

Related papers

• DeepGaze3.5-VL: Modeling Scanpaths via Autoregressive Token Prediction [21.818459524136376]
  DeepGaze3.5-VL establishes a new state-of-the-art across multiple datasets, achieving 2.18 bits of Information Gain (IG) on MIT1003, a 46% improvement over DeepGaze III.<n>Our generative framework serves as a powerful computational tool for direct behavioral interventions, allowing for controlled in-silico simulations that would be experimentally difficult or impossible to conduct in vivo.
  arXiv  Detail & Related papers  (2026-07-02T12:25:45Z)
• AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding [55.535374902204296]
  AffordanceVLA is a unified framework that introduces structured affordance forecasting as a task-oriented intermediate representation.<n>We show that AffordanceVLA achieves strong performance across diverse manipulation scenarios.
  arXiv  Detail & Related papers  (2026-06-04T13:28:51Z)
• Embodied3DBench: Benchmarking Low-Level Embodied Spatial Intelligence of Vision Language Models [50.6415287154632]
  Embodied3DBench is a robot-centric benchmark targeting low-level spatial intelligence in 3D environments.<n>The benchmark spans 12 subcategories and contains over 21k high-quality question-answer pairs.
  arXiv  Detail & Related papers  (2026-05-27T20:28:56Z)
• GEM: Generative Supervision Helps Embodied Intelligence [60.28190143133428]
  Embodied Vision-Language Models (VLMs) have demonstrated impressive performance and generalization in robotics.<n>A significant gap remains between the high-level semantic focus of standard text-guided pre-training paradigms and the low-level spatial and physical knowledge critical for execution in embodied environments.<n>We introduce GEM, a Generative-supervised Embodied vision-language Model designed to bridge this divide.
  arXiv  Detail & Related papers  (2026-05-27T14:39:42Z)
• OmniGF: A Dual-Branch Vision-Language Framework for Unified Gaze Following [59.53720386342017]
  We propose a unified vision-language framework that adapts foundational Vision-Language Models for scalable multi-person gaze reasoning.<n>By modeling all individuals, OmniGF seamlessly integrates precise spatial gaze target estimation, semantic gaze prediction, and complex social gaze reasoning.
  arXiv  Detail & Related papers  (2026-05-26T00:08:06Z)
• 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)
• VLADriver-RAG: Retrieval-Augmented Vision-Language-Action Models for Autonomous Driving [8.228888952282917]
  We propose textbfVLA-DriverRAG, a framework that grounds in explicit, structure-aware historical knowledge.<n>We abstract sensory inputs into semantic graphs via a textitVisual-to-Scenario mechanism, effectively filtering visual noise.<n>Experiments on the Bench2Drive benchmark establish a new state-of-the-art, achieving a Driving Score of 89.12.
  arXiv  Detail & Related papers  (2026-05-01T05:50:00Z)
• AULLM++: Structural Reasoning with Large Language Models for Micro-Expression Recognition [30.62262381459829]
  AULLM++ is a reasoning-oriented framework leveraging Large Language Models (LLMs)<n>It formulates AU prediction into three stages: evidence construction, structure modeling, and deduction-based prediction.<n>Experiments demonstrate AULLM++ achieves state-of-the-art performance on standard benchmarks.
  arXiv  Detail & Related papers  (2026-03-09T13:45:21Z)
• TagaVLM: Topology-Aware Global Action Reasoning for Vision-Language Navigation [70.23578202012048]
  Vision-Language Navigation (VLN) presents a unique challenge for Large Vision-Language Models (VLMs) due to their inherent architectural mismatch.<n>We propose TagaVLM (Topology-Aware Global Action reasoning), an end-to-end framework that explicitly injects topological structures into the VLM backbone.<n>To enhance topological node information, an Interleaved Navigation Prompt strengthens node-level visual-text alignment.<n>With the embedded topological graph, the model is capable of global action reasoning, allowing for robust path correction.
  arXiv  Detail & Related papers  (2026-03-03T13:28:07Z)
• Towards 3D Object-Centric Feature Learning for Semantic Scene Completion [18.41627244498394]
  Vision-based 3D Semantic Scene Completion (SSC) has received growing attention due to its potential in autonomous driving.<n>We propose Ocean, an object-centric prediction framework that decomposes the scene into individual object instances.<n>We show that Ocean achieves state-of-the-art performance, with mIoU scores of 17.40 and 20.28, respectively.
  arXiv  Detail & Related papers  (2025-11-17T06:28:26Z)
• FSDAM: Few-Shot Driving Attention Modeling via Vision-Language Coupling [5.609178055761294]
  We present FSDAM, a framework that achieves joint attention prediction and caption generation with 100 annotated examples.<n> FSDAM achieves competitive performance on attention prediction, generates coherent, and context-aware explanations.<n>This work shows that effective attention-conditioned generation is achievable with limited supervision, opening new possibilities for practical deployment of explainable driver attention systems.
  arXiv  Detail & Related papers  (2025-11-16T17:50:30Z)
• Enhancing 3D Gaze Estimation in the Wild using Weak Supervision with Gaze Following Labels [10.827081942898506]
  We introduce a novel Self-Training Weakly-Supervised Gaze Estimation framework (ST-WSGE)<n>We propose the Gaze Transformer (GaT), a modality-agnostic architecture capable of simultaneously learning static and dynamic gaze information from both image and video datasets.<n>By combining 3D video datasets with 2D gaze target labels from gaze following tasks, our approach achieves the following key contributions.
  arXiv  Detail & Related papers  (2025-02-27T16:35:25Z)
• DetCLIPv3: Towards Versatile Generative Open-vocabulary Object Detection [111.68263493302499]
  We introduce DetCLIPv3, a high-performing detector that excels at both open-vocabulary object detection and hierarchical labels. DetCLIPv3 is characterized by three core designs: 1) Versatile model architecture; 2) High information density data; and 3) Efficient training strategy. DetCLIPv3 demonstrates superior open-vocabulary detection performance, outperforming GLIPv2, GroundingDINO, and DetCLIPv2 by 18.0/19.6/6.6 AP, respectively.
  arXiv  Detail & Related papers  (2024-04-14T11:01:44Z)
• Towards Unified 3D Object Detection via Algorithm and Data Unification [70.27631528933482]
  We build the first unified multi-modal 3D object detection benchmark MM- Omni3D and extend the aforementioned monocular detector to its multi-modal version. We name the designed monocular and multi-modal detectors as UniMODE and MM-UniMODE, respectively.
  arXiv  Detail & Related papers  (2024-02-28T18:59:31Z)
• Unified Visual Relationship Detection with Vision and Language Models [89.77838890788638]
  This work focuses on training a single visual relationship detector predicting over the union of label spaces from multiple datasets. We propose UniVRD, a novel bottom-up method for Unified Visual Relationship Detection by leveraging vision and language models. Empirical results on both human-object interaction detection and scene-graph generation demonstrate the competitive performance of our model.
  arXiv  Detail & Related papers  (2023-03-16T00:06:28Z)

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.