mlsanityData to 7 Oct 2026

MiKASA: Multi-Key-Anchor & Scene-Aware Transformer for 3D Visual Grounding

arXiv 2403.03077 ↗First posted 5 Mar 2024cs.CVcited by 29

Not on the map. Why?

Abstract (arXiv)

3D visual grounding involves matching natural language descriptions with their corresponding objects in 3D spaces. Existing methods often face challenges with accuracy in object recognition and struggle in interpreting complex linguistic queries, particularly with descriptions that involve multiple anchors or are view-dependent. In response, we present the MiKASA (Multi-Key-Anchor Scene-Aware) Transformer. Our novel end-to-end trained model integrates a self-attention-based scene-aware object encoder and an original multi-key-anchor technique, enhancing object recognition accuracy and the understanding of spatial relationships. Furthermore, MiKASA improves the explainability of decision-making, facilitating error diagnosis. Our model achieves the highest overall accuracy in the Referit3D challenge for both the Sr3D and Nr3D datasets, particularly excelling by a large margin in categories that require viewpoint-dependent descriptions.

Key facts. Cited by 29 papers in the map (data through 7 Oct 2026). Builds on 25 of its 41 references that are in the map. First posted on arXiv 5 Mar 2024.

Builds on

25 of 41 references.

References · 25

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  5. 2017Graph Attention Networkscited by 8,854
  6. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  7. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  8. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  9. 2015Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Modelscited by 1,352
  10. 2016Modeling Context in Referring Expressionscited by 1,054
  11. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 432
  12. 2023Multi3DRefer: Grounding Text Description to Multiple 3D Objectscited by 142
13 more
  1. 2021InstanceRefer: Cooperative Holistic Understanding for Visual Grounding on Point Clouds through Instance Multi-level Contextual Referringcited by 140
  2. 2022Multi-View Transformer for 3D Visual Groundingcited by 138
  3. 2021Bottom Up Top Down Detection Transformers for Language Grounding in Images and Point Cloudscited by 128
  4. 2021SAT: 2D Semantics Assisted Training for 3D Visual Groundingcited by 124
  5. 2016Graph-Structured Representations for Visual Question Answeringcited by 123
  6. 2019Multimodal Intelligence: Representation Learning, Information Fusion, and Applicationscited by 89
  7. 2018CrisisMMD: Multimodal Twitter Datasets from Natural Disasterscited by 88
  8. 2021TransRefer3D: Entity-and-Relation Aware Transformer for Fine-Grained 3D Visual Groundingcited by 86
  9. 2023ViewRefer: Grasp the Multi-view Knowledge for 3D Visual Grounding with GPT and Prototype Guidancecited by 70
  10. 2018Structure Inference Net: Object Detection Using Scene-Level Context and Instance-Level Relationshipscited by 41
  11. 2022Look Around and Refer: 2D Synthetic Semantics Knowledge Distillation for 3D Visual Groundingcited by 41
  12. 2021Point Cloud Augmentation with Weighted Local Transformationscited by 39
  13. 2019Putting visual object recognition in contextcited by 33

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 15; 2026: 9.

29 papers.

Sort:

By area:

  1. 2024SeeGround: See and Ground for Zero-Shot Open-Vocabulary 3D Visual Groundingcited by 66
  2. 2024A Survey on Text-guided 3D Visual Grounding: Elements, Recent Advances, and Future Directionscited by 24
  3. 2025Unifying 2D and 3D Vision-Language Understandingcited by 15
  4. 2025DenseGrounding: Improving Dense Language-Vision Semantics for Ego-Centric 3D Visual Groundingcited by 12
  5. 2025SPAZER: Spatial-Semantic Progressive Reasoning Agent for Zero-shot 3D Visual Groundingcited by 12
  6. 2024Task-oriented Sequential Grounding and Navigation in 3D Scenescited by 9
  7. 2025ViewSRD: 3D Visual Grounding via Structured Multi-View Decompositioncited by 8
  8. 2025Inpaint360GS: Efficient Object-Aware 3D Inpainting via Gaussian Splatting for 360deg Scenescited by 8
  9. 20243D Spatial Understanding in MLLMs: Disambiguation and Evaluationcited by 5
  10. 2025ProxyTransformation: Preshaping Point Cloud Manifold With Proxy Attention For 3D Visual Groundingcited by 5
  11. 2024Multi-Object 3D Grounding with Dynamic Modules and Language-Informed Spatial Attentioncited by 4
  12. 2025From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenescited by 4
17 more
  1. 2025DEGround: An Effective Baseline for Ego-centric 3D Visual Grounding with a Homogeneous Frameworkcited by 4
  2. 2025GroundFlow: A Plug-in Module for Temporal Reasoning on 3D Point Cloud Sequential Groundingcited by 4
  3. 2025Seeing Clearly, Forgetting Deeply: Revisiting Fine-Tuned Video Generators for Driving Simulationcited by 3
  4. 2025PanoGrounder: Bridging 2D and 3D with Panoramic Scene Representations for VLM-based 3D Visual Groundingcited by 3
  5. 2025Zero-Shot 3D Visual Grounding from Vision-Language Modelscited by 2
  6. 2025TriCLIP-3D: A Unified Parameter-Efficient Framework for Tri-Modal 3D Visual Grounding based on CLIPcited by 2
  7. 2026Z3D: Zero-Shot 3D Visual Grounding from Imagescited by 2
  8. 2026Probing the Reliability of Driving VLMs: From Inconsistent Responses to Grounded Temporal Reasoningcited by 2
  9. 2026JOPP-3D: Joint Open Vocabulary Semantic Segmentation on Point Clouds and Panoramascited by 1
  10. 2025LSVG: Language-Guided Scene Graphs with 2D-Assisted Multi-Modal Encoding for 3D Visual Groundingcited by 0
  11. 2025B2N3D: Progressive Learning from Binary to N-ary Relationships for 3D Object Groundingcited by 0
  12. 2026Invaria: Learning Scale and Density Invariance in Point Clouds via Next-Resolution Predictioncited by 0
  13. 2026AgentGrounder: Zero-Shot 3D Visual Pointcloud Grounding using Multimodal Language Modelscited by 0
  14. 2026SSR3D-LLM: Structured Spatial Reasoning via Latent Steps for Fine-Grained Grounding in Unified 3D-LLMscited by 0
  15. 2026Agentic Collaborative Cognition for Zero-Shot 3D Understandingcited by 0
  16. 2026PruneGround: Plug-and-play Spatial Pruning for 3D Visual Groundingcited by 0
  17. 2026Where to Look Matters: Learning Influential Views for VLM-based 3D Visual Groundingcited by 0