mlsanityData to 14 Oct 2026

MOPA: Modular Object Navigation with PointGoal Agents

arXiv 2304.03696 ↗First posted 7 Apr 2023cs.RO cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

We propose a simple but effective modular approach MOPA (Modular ObjectNav with PointGoal agents) to systematically investigate the inherent modularity of the object navigation task in Embodied AI. MOPA consists of four modules: (a) an object detection module trained to identify objects from RGB images, (b) a map building module to build a semantic map of the observed objects, (c) an exploration module enabling the agent to explore the environment, and (d) a navigation module to move to identified target objects. We show that we can effectively reuse a pretrained PointGoal agent as the navigation model instead of learning to navigate from scratch, thus saving time and compute. We also compare various exploration strategies for MOPA and find that a simple uniform strategy significantly outperforms more advanced exploration methods.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 64 of its 80 references that are in the map. First posted on arXiv 7 Apr 2023.

Builds on

64 of 80 references.

References · 64

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  4. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,376
  5. 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,194
  6. 2019Habitat: A Platform for Embodied AI Researchcited by 1,191
  7. 2017Vision-and-Language Navigation: Interpreting visually-grounded navigation instructions in real environmentscited by 1,027
  8. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 869
  9. 2020Transformer Interpretability Beyond Attention Visualizationcited by 627
  10. 2022Detecting Twenty-thousand Classes using Image-level Supervisioncited by 614
  11. 2021Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AIcited by 592
  12. 2018On Evaluation of Embodied Navigation Agentscited by 547
38 more
  1. 2021Habitat 2.0: Training Home Assistants to Rearrange their Habitatcited by 511
  2. 2018Gibson Env: Real-World Perception for Embodied Agentscited by 481
  3. 2020Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environmentscited by 441
  4. 2020Object Goal Navigation using Goal-Oriented Semantic Explorationcited by 429
  5. 2015Neural Module Networkscited by 376
  6. 2019DD-PPO: Learning Near-Perfect PointGoal Navigators from 2.5 Billion Framescited by 362
  7. 2020Learning to Explore using Active Neural SLAMcited by 316
  8. 2019Self-Supervised Exploration via Disagreementcited by 265
  9. 2020ObjectNav Revisited: On Evaluation of Embodied Agents Navigating to Objectscited by 244
  10. 2018Touchdown: Natural Language Navigation and Spatial Reasoning in Visual Street Environmentscited by 239
  11. 2017Cognitive Mapping and Planning for Visual Navigationcited by 212
  12. 2020Neural Topological SLAM for Visual Navigationcited by 195
  13. 2022CoWs on Pasture: Baselines and Benchmarks for Language-Driven Zero-Shot Object Navigationcited by 194
  14. 2021Simple but Effective: CLIP Embeddings for Embodied AIcited by 191
  15. 2022ZSON: Zero-Shot Object-Goal Navigation using Multimodal Goal Embeddingscited by 178
  16. 2021The Surprising Effectiveness of Representation Learning for Visual Imitationcited by 154
  17. 2019Learning Exploration Policies for Navigationcited by 116
  18. 2022Habitat-Matterport 3D Semantics Datasetcited by 110
  19. 2020Occupancy Anticipation for Efficient Exploration and Navigationcited by 106
  20. 2022Navigating to Objects in the Real Worldcited by 106
  21. 2018Mapping Instructions to Actions in 3D Environments with Visual Goal Predictioncited by 101
  22. 2020MultiON: Benchmarking Semantic Map Memory using Multi-Object Navigationcited by 95
  23. 2021Language-Aligned Waypoint (LAW) Supervision for Vision-and-Language Navigation in Continuous Environmentscited by 94
  24. 2023PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNavcited by 93
  25. 2017Revisiting IM2GPS in the Deep Learning Eracited by 80
  26. 2022Offline Visual Representation Learning for Embodied Navigationcited by 75
  27. 2021Auxiliary Tasks and Exploration Enable ObjectNavcited by 74
  28. 2023OVRL-V2: A simple state-of-art baseline for ImageNav and ObjectNavcited by 70
  29. 2021No RL, No Simulation: Learning to Navigate without Navigatingcited by 65
  30. 2020An Exploration of Embodied Visual Explorationcited by 58
  31. 2020Semantic MapNet: Building Allocentric Semantic Maps and Representations from Egocentric Viewscited by 58
  32. 2018Visual Coreference Resolution in Visual Dialog using Neural Module Networkscited by 54
  33. 2020Auxiliary Tasks Speed Up Learning PointGoal Navigationcited by 54
  34. 2020AllenAct: A Framework for Embodied AI Researchcited by 54
  35. 2019Combining Optimal Control and Learning for Visual Navigation in Novel Environmentscited by 52
  36. 2021Drop the GAN: In Defense of Patches Nearest Neighbors as Single Image Generative Modelscited by 51
  37. 2021Learning to Map for Active Semantic Goal Navigationcited by 50
  38. 2022Retrospectives on the Embodied AI Workshopcited by 46

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 5; 2025: 4; 2026: 2.

11 papers.

  1. 2024EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understandingcited by 45
  2. 2024TANGO: Training-free Embodied AI Agents for Open-world Taskscited by 27
  3. 2024Perception Matters: Enhancing Embodied AI with Uncertainty-Aware Semantic Segmentationcited by 6
  4. 2026SpaMEM: Benchmarking Dynamic Spatial Reasoning via Perception-Memory Integration in Embodied Environmentscited by 6
  5. 2024Following the Human Thread in Social Navigationcited by 4
  6. 2024Zero-shot Object-Centric Instruction Following: Integrating Foundation Models with Traditional Navigationcited by 4
  7. 2025What does really matter in image goal navigation?cited by 2
  8. 2025MLFM: Multi-Layered Feature Maps for Richer Language Understanding in Zero-Shot Semantic Navigationcited by 1
  9. 2025DIV-Nav: Open-Vocabulary Spatial Relationships for Multi-Object Navigationcited by 1
  10. 2026EvoMemNav: Efficient Self-Evolving Fine-Grained Memory for Zero-Shot Embodied Navigationcited by 1
  11. 2025SEMNAV: Enhancing Visual Semantic Navigation in Robotics through Semantic Segmentationcited by 0