mlsanityData to 14 Oct 2026

Embodied Domain Adaptation for Object Detection

arXiv 2506.21860 ↗First posted 27 Jun 2025cs.RO cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Mobile robots rely on object detectors for perception and object localization in indoor environments. However, standard closed-set methods struggle to handle the diverse objects and dynamic conditions encountered in real homes and labs. Open-vocabulary object detection (OVOD), driven by Vision Language Models (VLMs), extends beyond fixed labels but still struggles with domain shifts in indoor environments. We introduce a Source-Free Domain Adaptation (SFDA) approach that adapts a pre-trained model without accessing source data. We refine pseudo labels via temporal clustering, employ multi-scale threshold fusion, and apply a Mean Teacher framework with contrastive learning. Our Embodied Domain Adaptation for Object Detection (EDAOD) benchmark evaluates adaptation under sequential changes in lighting, layout, and object diversity. Our experiments show significant gains in zero-shot detection performance and flexible adaptation to dynamic indoor conditions.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 38 of its 43 references that are in the map. First posted on arXiv 27 Jun 2025.

Builds on

38 of 43 references.

References · 38

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  5. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  6. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,444
  7. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,332
  8. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  9. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,376
  10. 2015Learning Transferable Features with Deep Adaptation Networkscited by 1,329
  11. 2017Cascade R-CNN: Delving into High Quality Object Detectioncited by 1,276
  12. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 869
26 more
  1. 2022Detecting Twenty-thousand Classes using Image-level Supervisioncited by 614
  2. 2024YOLO-World: Real-Time Open-Vocabulary Object Detectioncited by 503
  3. 2018Domain Adaptive Faster R-CNN for Object Detection in the Wildcited by 443
  4. 2017Semantic Foggy Scene Understanding with Synthetic Datacited by 403
  5. 2019Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Comingcited by 336
  6. 2021Unbiased Teacher for Semi-Supervised Object Detectioncited by 306
  7. 2018Strong-Weak Distribution Alignment for Adaptive Object Detectioncited by 270
  8. 2018Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptationcited by 218
  9. 2016Driving in the Matrix: Can Virtual Worlds Replace Human-Generated Annotations for Real World Tasks?cited by 194
  10. 2021Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Datacited by 152
  11. 2022SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptationcited by 146
  12. 2021Probabilistic two-stage detectioncited by 130
  13. 2020A Free Lunch for Unsupervised Domain Adaptive Object Detection without Source Datacited by 79
  14. 2020Semantic MapNet: Building Allocentric Semantic Maps and Representations from Egocentric Viewscited by 58
  15. 2022Instance Relation Graph Guided Source-Free Domain Adaptive Object Detectioncited by 55
  16. 2022Towards Online Domain Adaptive Object Detectioncited by 36
  17. 2022ConfMix: Unsupervised Domain Adaptation for Object Detection via Confidence-based Mixingcited by 28
  18. 2022Interactron: Embodied Adaptive Object Detectioncited by 21
  19. 2018Pushing the Limits of Unconstrained Face Detection: a Challenge Dataset and Baseline Resultscited by 16
  20. 2023Evidential Active Recognition: Intelligent and Prudent Open-World Embodied Perceptioncited by 15
  21. 2024Simplifying Source-Free Domain Adaptation for Object Detection: Effective Self-Training Strategies and Performance Insightscited by 14
  22. 2023Active Open-Vocabulary Recognition: Let Intelligent Moving Mitigate CLIP Limitationscited by 6
  23. 2024Domain Adaptation for Large-Vocabulary Object Detectorscited by 2
  24. 2024Enhancing Embodied Object Detection through Language-Image Pre-training and Implicit Object Memorycited by 2
  25. 2023Improving Online Source-free Domain Adaptation for Object Detection by Unsupervised Data Acquisitioncited by 1
  26. 2024Source-Free Unsupervised Domain Adaptation with Hypothesis Consolidation of Prediction Rationalecited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Towards Robust Cross-Dataset Object Detection Generalization under Domain Specificitycited by 0
  2. 2026See, Act, Adapt: Active Perception for Unsupervised Cross-Domain Visual Adaptation via Personalized VLM-Guided Agentcited by 0
  3. 2026Embodied Active Learning under Limited Annotation and Navigation Budget for Object Detectioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025UPRE: Zero-Shot Domain Adaptation for Object Detection via Unified Prompt and Representation Enhancement
  2. 2026The Detector Teaches Itself: Lightweight Self-Supervised Adaptation for Open-Vocabulary Object Detection
  3. 2024Feature Based Methods in Domain Adaptation for Object Detection: A Review Paper
  4. 2025Textual Inversion for Efficient Adaptation of Open-Vocabulary Object Detectors Without Forgetting
  5. 2025Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Models
  6. 2025Bayesian Test-time Adaptation for Object Recognition and Detection with Vision-language Models
  7. 2024DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detection
  8. 2026Learning Multi-Modal Prototypes for Cross-Domain Few-Shot Object Detection