Pith. sign in

REVIEW 3 major objections 5 minor 3 cited by

Semantic Mapping in Indoor Embodied AI -- A Survey on Advances, Challenges, and Future Directions

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A survey sorts indoor robot semantic maps on two independent axes: structure and encoding.

desk verdict A useful, well-organized survey whose two-axis taxonomy is sensible and current, but whose trend claim rests on a curated corpus with no stated search methodology. read the letter →

arxiv 2501.05750 v3 pith:5VN2OLFE submitted 2025-01-10 cs.RO cs.CV

Sonia Raychaudhuri , Angel X. Chang This is my paper · ORCID
classification cs.ROcs.CV
keywords semanticmappingembodiedAIindoornavigationmapstructureencodingopen-vocabularymapssurveytaxonomyroboticperception
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that every semantic-map-building method in indoor embodied AI can be understood as a combination of two independent design choices: the structure of the map (spatial grid, topological graph, dense geometric, or hybrid) and the encoding stored in it (explicit labels or implicit learned features). It reviews the literature through this lens and claims the field is converging on open-vocabulary, queryable, task-agnostic maps, while memory demand and computational inefficiency remain open problems. A sympathetic reader would care because the taxonomy turns a scattered body of navigation, exploration, and manipulation papers into comparable categories, and the identified trend indicates where future effort is likely to pay off.

What carries the argument

The organizing device is the two-axis taxonomy itself: structure versus encoding. Structure determines how locations and landmarks are stored and how well the map scales; encoding determines what can be queried and whether the map can handle unseen categories. Works that project encoded features and then decode them into explicit labels are treated as an intermediate 'implicit to explicit' category, and evaluation is split into extrinsic task-level metrics versus intrinsic map-quality metrics across accuracy, completeness, consistency, and robustness, which the paper argues are underdeveloped.

What would settle it

A reader could test the trend claim by searching broadly for intrinsic map-evaluation papers published before 2020; if many such papers exist in robotics or SLAM venues that the survey does not cover, the claim that intrinsic evaluation is largely missing would need revision.

Watch

Extended reading notes

Core claim

The central claim is that map representation, not the downstream task, is the right organizing principle for semantic mapping research. The authors classify surveyed methods along two axes: map structure, which covers spatial grids that store metric information cell-by-cell, topological graphs that store landmark nodes and edges, dense geometric maps that attach semantics to point clouds, meshes, surfels, or neural fields, and hybrid maps that combine several of these; and semantic encoding, which covers explicit values such as occupancy, object category, room type, exploration state, and audio intensity, versus implicit features from pretrained encoders that are closed-vocabulary when trained on fixed categories and open-vocabulary when derived from vision-language models. The paper asserts that recent work is moving toward open-vocabulary, queryable maps that can be built once and reused across tasks, and that the main bottlenecks are memory demands and computational inefficiency.

Load-bearing premise

The survey's conclusions assume the papers it selected are representative of all semantic-map-building work in indoor embodied AI, since no systematic search or inclusion criteria are given.

Editorial extensions

If this is right

  • If the taxonomy is right, method comparisons should first fix the structure-encoding pair, since different pairs have different scaling, querying, and memory properties.
  • If the open-vocabulary trend holds, future maps will be built once and queried by arbitrary natural language, reducing the need for task-specific retraining.
  • If intrinsic evaluation remains neglected, claims that a method improved task success will stay ambiguous about whether the map itself got better.
  • If memory and compute bottlenecks persist, dense and open-vocabulary maps will remain limited to offline or simulated settings until more efficient representations appear.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same two-axis lens could be extended to outdoor mapping, where bird's-eye-view representations in autonomous driving resemble spatial grids with implicit encodings, a connection the survey mentions only in passing.
  • A direct test of the taxonomy would be a benchmark that holds the downstream task fixed and varies only map structure or only encoding, isolating each axis' contribution to query accuracy, memory use, and navigation success.
  • Because the survey does not report a systematic search or inclusion criteria, its trend claims are conditional on the selected corpus; a broader search of robotics and SLAM venues could shift the balance of papers and the apparent direction of the field.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This survey reviews semantic map-building methods for indoor embodied agents. It organizes the literature along two axes: map structure (spatial grids, topological graphs, dense geometric maps, and hybrids) and semantic encoding (explicit annotations versus implicit features, further split into closed- and open-vocabulary). The paper also provides background on embodied tasks and SLAM, summarizes the map-building pipeline, discusses evaluation practices, and lists open challenges and future directions. Its central claims are that the proposed taxonomy is a useful unifying organizational scheme and that the field is moving toward open-vocabulary, queryable, task-agnostic map representations, while memory and compute efficiency remain bottlenecks.

Significance. If the reviewed corpus is representative, this is a useful and timely survey. Its two-axis taxonomy is intuitive and does organize many influential methods, from classical SLAM-based semantic mapping to recent CLIP-based open-vocabulary maps. The survey also bridges robotics and embodied AI, provides structured tables for method comparison, and highlights the underdevelopment of intrinsic map evaluation, which is a real gap. The paper is best when it is descriptive: the per-method summaries in Sections 4 and 5 are broadly consistent with the cited literature, and the discussion of trade-offs among map structures is balanced. The central trend claim, however, is anchored to a manually assembled corpus without documented selection criteria, so the survey's forward-looking conclusions are less secure than its taxonomy.

major comments (3)
  1. [Section 1, Tables 1–6, Fig. 6] The survey provides no search strategy, database list, query terms, date range, or inclusion/exclusion criteria for the reviewed works. The trend claim that 'the last few years have seen a shift toward open-vocabulary semantic maps' (Fig. 6 caption; Sec. 8; abstract) depends on the corpus being representative, but the timeline is manually constructed and the open-vocabulary papers in Table 6 are not compared against the full set of semantic-mapping publications per year. Without a reproducible corpus, the trend could be an artifact of selecting recent CLIP-based works. Please specify the literature search methodology and report quantitative coverage statistics (e.g., number of papers screened, included, and per-year counts for each encoding type).
  2. [Section 6.2, Table 7] The paper claims that intrinsic map evaluation is 'very little' studied and that 'none of the prior works measure semantic consistency,' yet Section 6.2 itself describes prior intrinsic evaluations (e.g., SemanticMapNet, OpenScene, ConceptGraphs, OpenLex3D) and Table 7 lists consistency metrics from the SLAM literature. The qualitative claim may be defensible if 'little' means 'no standardized suite,' but the current wording overstates the gap and is contradicted by the paper's own evidence. Please sharpen the claim to 'no standardized intrinsic evaluation framework exists' and either substantiate or qualify the assertion that semantic consistency is never measured.
  3. [Section 8.1, Section 5.2.2, Table 6] The paper repeatedly describes open-vocabulary maps as 'task-agnostic' and 'general-purpose,' but most of the reviewed evidence is task-specific: Table 6 lists works evaluated on ObjectNav, VLN, manipulation, or scene understanding, with only a few (e.g., ConceptGraphs) demonstrating the same map across multiple downstream tasks. As written, 'task-agnostic' is an aspiration rather than an established property. Please define the term precisely and provide direct evidence for multi-task reuse from the surveyed works, or relax the claim accordingly.
minor comments (5)
  1. [Abstract] The abstract contains typographical errors: 'spatialgrids,' 'densegeometric,' 'implicit features orexplicit,' and 'still remaining to be open challenges.' These should be corrected.
  2. [Table 1] The symbol legend (occupancy, explored-area, object category, visitation time) does not render consistently, with several cells showing raw glyphs like 'Ăˆ' and 'x.' This makes the central summary table hard to read; please use text labels or a clean legend.
  3. [Eq. (6) and (7)] There is a duplicated 'where where' and the projection matrices are not defined with enough care: P_v is called 'known orthographic projection matrix' but its dimensions are never specified. Please clarify the notation.
  4. [Fig. 6] The 'timeline' is presented as a static list of paper names grouped by period rather than a quantitative timeline. Consider replacing it with a proper chart that shows publication counts per structure/encoding over time, which would also support the paper's trend claim.
  5. [Section 5.1 and Table 1] Several works by the authors themselves (MOPA, LIFGIF, ASHiTA) are discussed favorably in the text and appear in the tables. This is not inherently inappropriate, but the absence of documented inclusion criteria makes it harder to rule out selection bias; the methodology suggested in the major comments would also address this concern.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the taxonomy and trend claims are organizing summaries of external prior work, not derivations from the paper's own assumptions.

full rationale

This is a survey. Its central contribution is a two-axis taxonomy (map structure x semantic encoding) and a trend observation (movement toward open-vocabulary, queryable maps). The classification criteria in Sections 3.2-3.3 are definitions, and the papers in Tables 1-6 are categorized by those definitions; this is not a derivation whose conclusion is assumed in its input. The trend claim in Section 4 / Figure 6 and Section 8 is an empirical summary of the reviewed timeline, not a theorem. The authors cite their own prior work (MOPA, Raychaudhuri et al. 2023; LIFGIF, Raychaudhuri et al. 2025) as examples in the taxonomy tables and modular-pipeline discussion, but these citations are descriptive and not load-bearing: MOPA and LIFGIF are explicit-encoding methods, and the open-vocabulary trend claim is supported by many independent works in Table 6. No uniqueness theorem, fitted parameter, or derived equation is invoked. The absence of a systematic search protocol is a corpus-completeness or selection concern, not circularity, because the survey's claims could in principle be falsified by a broader or differently sampled literature; there is no definitional or equation-level step that forces the conclusion.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

As a survey, the paper's load-bearing premises are the exhaustiveness of its taxonomy and the representativeness of its literature selection. These are domain assumptions rather than standard mathematical axioms. No free parameters or invented entities are introduced.

assumptions (2)
  • domain assumption Semantic map-building methods can be exhaustively partitioned by structure (grid, topological, dense geometric, hybrid) and encoding (explicit, implicit).
    This is the central organizing premise of the survey, asserted in Section 1 and Figure 1 without proof that the categories are mutually exclusive and jointly exhaustive.
  • domain assumption The reviewed papers are representative of the field of indoor semantic mapping in embodied AI.
    The survey claims comprehensiveness but does not describe its literature search or inclusion criteria, so the representativeness of Tables 1 to 6 is assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Semantic Mapping in Indoor Embodied AI -- A Survey on Advances, Challenges, and Future Directions." pith.science (2026). https://pith.science/paper/5VN2OLFE

@misc{pith2026250105750,
  author       = {Pith},
  title        = {Pith review of: Semantic Mapping in Indoor Embodied AI -- A Survey on Advances, Challenges, and Future Directions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5VN2OLFE}},
  note         = {Machine review of arXiv:2501.05750}
}
read the original abstract

Intelligent embodied agents (e.g. robots) need to perform complex semantic tasks in unfamiliar environments. Among many skills that the agents need to possess, building and maintaining a semantic map of the environment is most crucial in long-horizon tasks. A semantic map captures information about the environment in a structured way, allowing the agent to reference it for advanced reasoning throughout the task. While existing surveys in embodied AI focus on general advancements or specific tasks like navigation and manipulation, this paper provides a comprehensive review of semantic map-building approaches in embodied AI, specifically for indoor navigation. We categorize these approaches based on their structural representation (spatial grids, topological graphs, dense point-clouds or hybrid maps) and the type of information they encode (implicit features or explicit environmental data). We also explore the strengths and limitations of the map building techniques, highlight current challenges, and propose future research directions. We identify that the field is moving towards developing open-vocabulary, queryable, task-agnostic map representations, while high memory demands and computational inefficiency still remaining to be open challenges. This survey aims to guide current and future researchers in advancing semantic mapping techniques for embodied AI systems.

Figures

Figures reproduced from arXiv: 2501.05750 by the authors.

Figure 1
Figure 1. Semantic maps. The survey categorizes semantic map building methods in embodied agents based on their structure and the encoding it stores. Structure: a map of a physical environment can be structured as a topological map (with nodes and edges), a spatial grid, a dense geometric map or a hybrid map combining two or more of the others. Encoding: the structured maps can store either explicit (occupancy, object type, e… view at source ↗
Figure 2
Figure 2. Motivation. To perform a complex task in an indoor environment, the robotic agent must possess multiple skills of language understanding, visual perception, navigation, etc. Among these the most crucial is building and maintaining a semantic map of the environment so that it can come back to it while performing the task. 2.1.1 Robotics tasks The era of modern robotics began with Unimate (Detesan & Moholea, 2024), th… view at source ↗
Figure 3
Figure 3. SLAM. (left) At the core of classical mobile robotics lie three core tasks – mapping, localization, and planning. These are often interdependent on each other and overlap to form other tasks such as SLAM (localization and mapping), exploration (mapping and planning), active localization (planning and localization) and active SLAM (mapping, localization, and planning) [figure reproduced from Fairfield (2009)]. (right… view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: End-to-end vs Modular. (Top) End-to-end model is trained as a single pipeline which generates actions directly from sensory inputs. (Bottom) Modular pipeline consists of various sub-modules, each with a specific function so that they can be trained independently of the…
Figure 5
Figure 5. Figure 5: Map building involves localization (where the agent is on the map), feature extraction (extracting useful semantic information from the observations), and map update (building the map by aggregating the semantic information over time). 3.3 What encoding is stored in th…
Figure 6
Figure 6. Figure 6: This timeline highlights how semantic mapping methods have progressed over the years, with [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: Grid map building. A spatial grid map has dimensions (M ˆ N ˆ K) where M and N are spatial dimensions and K is the number of semantic channels. The common pipeline to build the map is to segment the input image, then ground project into an egocentric map egomap, which …
Figure 8
Figure 8. Figure 8: Map Encoding refers to the values stored in the map and can be either explicit or implicit, depending on whether the information is hand-selected or a learned feature representation of the observation. 22 [PITH_FULL_IMAGE:figures/full_fig_p022_8.png]
Figure 9
Figure 9. Figure 9: Open-Vocabulary map building. There has been a growing interest to build flexible open￾vocabulary maps which can be built once and then used in various downstream tasks during inference. VLMap (Huang et al., 2023a) and NLMap (Chen et al., 2023a) structure their maps as…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Robotic Contextual Awareness for Human-Robot Collaboration and Environmental Understanding

    cs.RO 2026-07 conditional novelty 6.0 of 10

    Novel person re-identification with continual adaptation plus submap LiDAR SLAM, ground-aware filtering, Gaussian Scan Context, and multi-modal semantic mapping improve robotic contextual awareness for HRC and navigation.

  2. Map as a By-product: Collective Landmark Mapping from IMU Data and User-provided Texts in Situated Tasks

    cs.HC 2025-09 conditional novelty 5.0 of 10

    Indoor workers' routine text notes plus phone IMU data can be aggregated into accurate semantic landmark maps, with 0.8 m mean error in a retail store and 1.9 m in an office with 11 records.

  3. Large Model Empowered Embodied AI: A Survey on Decision-Making and Embodied Learning

    cs.RO 2025-08 reject novelty 4.0 of 10

    A review that categorizes large-model-empowered embodied AI into hierarchical and end-to-end decision-making, imitation and reinforcement learning, and world models.

Reference graph

Works this paper leans on

39 extracted references · 27 canonical work pages · cited by 3 Pith papers

  1. [4]

    3, 5, 8, 9, 10, 16, 23, 27, 29, 30 38 Devendra Singh Chaplot, Dhiraj Prakashchand Gandhi, Abhinav Gupta, and Russ R Salakhutdinov

    URL https://openreview.net/forum?id=HklXn1BKDH. 3, 5, 8, 9, 10, 16, 23, 27, 29, 30 38 Devendra Singh Chaplot, Dhiraj Prakashchand Gandhi, Abhinav Gupta, and Russ R Salakhutdinov. Object goal navigation using goal-oriented semantic exploration. InAdvances in Neural Information Processing Systems, volume 33, pp. 4247–4258, 2020a. URLhttps://proceedings.neur...

    work page arXiv 2020

  2. [6]

    3, 16, 23, 30 Georgios Georgakis, Bernadette Bucher, Anton Arapin, Karl Schmeckpeper, Nikolai Matni, and Kostas Daniilidis

    URL https://par.nsf.gov/servlets/purl/10340845. 3, 16, 23, 30 Georgios Georgakis, Bernadette Bucher, Anton Arapin, Karl Schmeckpeper, Nikolai Matni, and Kostas Daniilidis. Uncertainty-driven planner for exploration and navigation. InInternational Conference on Robotics and Automation (ICRA), pp. 11295–11302. IEEE, 2022a. URLhttps://ieeexplore.ieee.org/ ab...

    work page arXiv 2023

  3. [10]

    10, 26 Dong Wook Ko, Chuho Yi, and Il Hong Suh

    URL http://openaccess.thecvf.com/content/ICCV2023/html/Kirillov_Segment_Anything_ ICCV_2023_paper.html. 10, 26 Dong Wook Ko, Chuho Yi, and Il Hong Suh. Semantic mapping and navigation: A Bayesian approach. In IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 2630–2636. IEEE, 2013. URL https://ieeexplore.ieee.org/abstract/document/66...

    work page arXiv 2013

  4. [14]

    8 Yen-Chen Lin, Andy Zeng, Shuran Song, Phillip Isola, and Tsung-Yi Lin

    URL https://www.sciencedirect.com/science/article/pii/S2666651022000146. 8 Yen-Chen Lin, Andy Zeng, Shuran Song, Phillip Isola, and Tsung-Yi Lin. Learning to see before learning to act: Visual pre-training for manipulation. InIEEE International Conference on Robotics and Automation,

    work page

  5. [16]

    3 Pietro Mazzaglia, Ozan Catal, Tim Verbelen, and Bart Dhoedt

    URL https://ieeexplore.ieee.org/abstract/document/10160800/. 3 Pietro Mazzaglia, Ozan Catal, Tim Verbelen, and Bart Dhoedt. Curiosity-driven exploration via latent bayesian surprise. InProceedings of the AAAI conference on artificial intelligence, volume 36, pp. 7752–7760,

    work page arXiv

  6. [17]

    9 John McCormac, Ankur Handa, Andrew Davison, and Stefan Leutenegger

    URL https://ojs.aaai.org/index.php/AAAI/article/view/20743. 9 John McCormac, Ankur Handa, Andrew Davison, and Stefan Leutenegger. Semanticfusion: Dense 3d semantic mapping with convolutional neural networks. InIEEE International Conference on Robotics and Automation, pp. 4628–4635. IEEE, 2017. URLhttps://ieeexplore.ieee.org/abstract/document/7989538/. 3, ...

    work page arXiv 2017

  7. [18]

    8 Lars Mescheder, Michael Oechsle, Michael Niemeyer, Sebastian Nowozin, and Andreas Geiger

    URL https://ojs.aaai.org/index.php/AAAI/article/view/10364. 8 Lars Mescheder, Michael Oechsle, Michael Niemeyer, Sebastian Nowozin, and Andreas Geiger. Occupancy networks: Learning 3d reconstruction in function space. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 4460–4470, 2019. URL http://openaccess.thecvf.co...

    work page arXiv 2019

  8. [19]

    7 Zhixiang Min and Enrique Dunn

    URL https://ieeexplore.ieee.org/abstract/document/9830880/. 7 Zhixiang Min and Enrique Dunn. Voldor+ slam: For the times when feature-based or direct methods are not good enough. InIEEE International Conference on Robotics and Automation, pp. 13813–13819. IEEE,

    work page arXiv

Show all 39 references
  1. [20]

    3, 19 Oscar Martinez Mozos, Rudolph Triebel, Patric Jensfelt, Axel Rottmann, and Wolfram Burgard

    URL https://ieeexplore.ieee.org/abstract/document/9561230/. 3, 19 Oscar Martinez Mozos, Rudolph Triebel, Patric Jensfelt, Axel Rottmann, and Wolfram Burgard. Supervised se- manticlabelingofplacesusinginformationextractedfromsensordata. Robotics and Autonomous Systems,55 (5):39...

    2007

  2. [21]

    3, 10, 18, 19, 20, 28 Rolf Pfeifer and Fumiya Iida

    URL http://openaccess.thecvf.com/content/CVPR2023/html/Peng_OpenScene_3D_Scene_ Understanding_With_Open_Vocabularies_CVPR_2023_paper.html. 3, 10, 18, 19, 20, 28 Rolf Pfeifer and Fumiya Iida. Embodied artificial intelligence: Trends and challenges.Lecture notes in computer scie...

    2004 arXiv

  3. [22]

    29, 30 Andrzej Pronobis and Patric Jensfelt

    URL https://api.semanticscholar.org/CorpusID:195884278. 29, 30 Andrzej Pronobis and Patric Jensfelt. Large-scale semantic mapping and reasoning with heterogeneous modalities. In IEEE international conference on robotics and automation, pp. 3515–3522. IEEE, 2012. URL https://ie...

    2012

  4. [23]

    7 52 Zhentian Qian, Jie Fu, and Jing Xiao

    URL https://ieeexplore.ieee.org/abstract/document/9561532/. 7 52 Zhentian Qian, Jie Fu, and Jing Xiao. Towards Accurate Loop Closure Detection in Semantic SLAM With 3D Semantic Covisibility Graphs. IEEE Robotics and Automation Letters, 7:2455–2462, 2022. URL https://api.semant...

    2022

  5. [24]

    7 Minghan Qin, Wanhua Li, Jiawei Zhou, Haoqian Wang, and Hanspeter Pfister

    URL https://www.sciencedirect.com/science/article/pii/S1047320321000389. 7 Minghan Qin, Wanhua Li, Jiawei Zhou, Haoqian Wang, and Hanspeter Pfister. LangSplat: 3D language Gaussian splatting. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition,...

    2024 arXiv

  6. [25]

    27 Sonia Raychaudhuri, Tommaso Campari, Unnat Jain, Manolis Savva, and Angel X

    URL https://aclanthology.org/2021.emnlp-main.328/. 27 Sonia Raychaudhuri, Tommaso Campari, Unnat Jain, Manolis Savva, and Angel X. Chang. MOPA: Modular Object Navigation with PointGoal Agents. arXiv preprint arXiv:2304.03696, 2023. URL https:// 3dlg-hcvc.github.io/mopa/. 3, 8,...

    2021 arXiv

  7. [27]

    9 53 Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun

    URL https://arxiv.org/abs/2403.15941. 9 53 Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster R-CNN: Towards real-time object detection with region proposal networks.Advances in Neural Information Processing Systems, 28, 2015. URLhttps:// proceedings.neurips.cc/pape...

    2015 arXiv

  8. [28]

    19 Renato F Salas-Moreno, Richard A Newcombe, Hauke Strasdat, Paul HJ Kelly, and Andrew J Davi- son

    URL https://ieeexplore.ieee.org/abstract/document/7989518/. 19 Renato F Salas-Moreno, Richard A Newcombe, Hauke Strasdat, Paul HJ Kelly, and Andrew J Davi- son. Slam++: Simultaneous localisation and mapping at the level of objects. In Proceedings of the IEEE/CVF Conference on ...

    2013

  9. [29]

    5, 16 54 Thomas Schops, Torsten Sattler, and Marc Pollefeys

    URL http://openaccess.thecvf.com/content_ICCV_2019/html/Savva_Habitat_A_Platform_ for_Embodied_AI_Research_ICCV_2019_paper.html. 5, 16 54 Thomas Schops, Torsten Sattler, and Marc Pollefeys. Bad slam: Bundle adjusted direct rgb-d slam. In Proceedings of the IEEE/CVF Conference ...

    2019

  10. [31]

    3, 7, 16, 17, 18, 25, 26 Mo Shan, Qiaojun Feng, and Nikolay Atanasov

    URL https://openreview.net/forum?id=UW5A3SweAH. 3, 7, 16, 17, 18, 25, 26 Mo Shan, Qiaojun Feng, and Nikolay Atanasov. OrcVIO: Object residual constrained visual-inertial odometry. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 5104–5111. IEE...

    2023

  11. [33]

    3, 20 Sebastian Thrun

    URL http://openaccess.thecvf.com/content_ICCV_2019/html/Thomas_KPConv_Flexible_ and_Deformable_Convolution_for_Point_Clouds_ICCV_2019_paper.html. 3, 20 Sebastian Thrun. Probabilistic robotics. Communications of the ACM, 45(3):52–57, 2002. URL https: //dl.acm.org/doi/fullHtml/1...

    2002

  12. [34]

    3, 19 Konstantinos A Tsintotas, Loukas Bampis, and Antonios Gasteratos

    URL https://ojs.aaai.org/index.php/AAAI/article/view/8378. 3, 19 Konstantinos A Tsintotas, Loukas Bampis, and Antonios Gasteratos. The revisiting problem in simultaneous localization and mapping: A survey on visual loop closure detection.IEEE Transactions on Intelligent Transp...

    2022

  13. [36]

    16 Yu Xiang and Dieter Fox

    URL http://openaccess.thecvf.com/content_CVPR_2020/html/Xiang_SAPIEN_A_SimulAted_ Part-Based_Interactive_ENvironment_CVPR_2020_paper.html. 16 Yu Xiang and Dieter Fox. DA-RNN: Semantic mapping with data associated recurrent neural networks. arXiv preprint arXiv:1703.03098, 2017...

    2017 arXiv

  14. [37]

    Atlanta, GA, USA

    URL https://ieeexplore.ieee.org/abstract/document/10610712/. Atlanta, GA, USA. 3, 9, 25, 29 Naoki Yokoyama, Ram Ramrakhya, Abhishek Das, Dhruv Batra, and Sehoon Ha. HM3D-OVON: A Dataset and Benchmark for Open-Vocabulary Object Goal Navigation.arXiv preprint arXiv:2409.14296, 2...

    2024

  15. [38]

    3, 23 Tianyao Zhang, Xiaoguang Hu, Jin Xiao, and Guofeng Zhang

    URL https://arxiv.org/abs/2502.13451. 3, 23 Tianyao Zhang, Xiaoguang Hu, Jin Xiao, and Guofeng Zhang. A survey of visual navigation: From geometry to embodied AI.Engineering Applications of Artificial Intelligence, 114:105036, 2022b. URL https://ieeexplore.ieee.org/abstract/do...

    2021 arXiv

  16. [39]

    20 Junhui Zhao, Jingyue Shi, and Li Zhuo

    URL https://openaccess.thecvf.com/content/ICCV2021/html/Zhao_Point_Transformer_ ICCV_2021_paper.html. 20 Junhui Zhao, Jingyue Shi, and Li Zhuo. BEV perception for autonomous driving: State of the art and future perspectives. Expert Systems with Applications, 258:125103, 2024. ...

    2024 arXiv

  17. [179]

    URLhttps://link.springer.com/chapter/10.1007/978-3-031-72751-1_10

    Springer, 2024a. URLhttps://link.springer.com/chapter/10.1007/978-3-031-72751-1_10. 19, 20 47 Tianyu Li, Peijin Jia, Bangjun Wang, Li Chen, Kun Jiang, Junchi Yan, and Hongyang Li. LaneSegNet: Map Learning with Lane Segment Perception for Autonomous Driving. InInternational Con...

  18. [1996]

    1117/12.240962.short

    URL https://www.spiedigitallibrary.org/conference-proceedings-of-spie/2726/0000/ Fast-marching-level-set-methods-for-three-dimensional-photolithography-development/10. 1117/12.240962.short. 9 Nur Muhammad Mahi Shafiullah, Chris Paxton, Lerrel Pinto, Soumith Chintala, and Arthu...

    arXiv

  19. [2001]

    3, 17, 23 Javier Civera, Dorian Gálvez-López, Luis Riazuelo, Juan D Tardós, and Jose Maria Martinez Montiel

    URL https://ieeexplore.ieee.org/abstract/document/928558/. 3, 17, 23 Javier Civera, Dorian Gálvez-López, Luis Riazuelo, Juan D Tardós, and Jose Maria Martinez Montiel. Towards semantic SLAM using a monocular camera. InIEEE/RSJ international conference on intelligent robots and...

    2011

  20. [2012]

    3, 19 Muhammad Sualeh and Gon-Woo Kim

    URL https://ieeexplore.ieee.org/abstract/document/6385983/. 3, 19 Muhammad Sualeh and Gon-Woo Kim. Simultaneous localization and mapping in the epoch of semantics: a survey.International Journal of Control, Automation and Systems, 17(3):729–742, 2019. URLhttps: //link.springer...

    2019

  21. [2014]

    7 Obin Kwon, Nuri Kim, Yunho Choi, Hwiyeon Yoo, Jeongho Park, and Songhwai Oh

    URL https://link.springer.com/chapter/10.1007/978-3-319-10599-4_45. 7 Obin Kwon, Nuri Kim, Yunho Choi, Hwiyeon Yoo, Jeongho Park, and Songhwai Oh. Vi- sual graph memory with unsupervised representation for visual navigation. In Proceedings of the IEEE/CVF International Confere...

  22. [2015]

    19 Thomas Whelan, Renato F Salas-Moreno, Ben Glocker, Andrew J Davison, and Stefan Leutenegger

    URL https://roboticsproceedings.org/rss11/p01.pdf. 19 Thomas Whelan, Renato F Salas-Moreno, Ben Glocker, Andrew J Davison, and Stefan Leutenegger. Elastic- Fusion: Real-time dense SLAM and light source estimation.The International Journal of Robotics Research, 35(14):1697–1716...

    2016 arXiv

  23. [2016]

    9, 10, 15, 17 Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick

    URL http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_ CVPR_2016_paper.html. 9, 10, 15, 17 Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick. Mask R-CNN. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 296...

    2011

  24. [2017]

    7, 18 Finn L Busch, Timon Homberger, Jesús Ortega-Peimbert, Quantao Yang, and Olov Andersson

    URL https://ieeexplore.ieee.org/abstract/document/7989203/. 7, 18 Finn L Busch, Timon Homberger, Jesús Ortega-Peimbert, Quantao Yang, and Olov Andersson. One Map to Find Them All: Real-time Open-Vocabulary Mapping for Zero-shot Multi-Object Navigation. InIEEE International Con...

    2025

  25. [2019]

    3, 21 Nikolay Atanasov, Sean L Bowman, Kostas Daniilidis, and George J Pappas

    URL http://openaccess.thecvf.com/content_ICCV_2019/html/Armeni_3D_Scene_Graph_A_ Structure_for_Unified_Semantics_3D_Space_ICCV_2019_paper.html. 3, 21 Nikolay Atanasov, Sean L Bowman, Kostas Daniilidis, and George J Pappas. A Unifying View of Geometry, Semantics, and Data Assoc...

    2018

  26. [2020]

    28 Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee

    URL https://ieeexplore.ieee.org/abstract/document/9197331/. 28 Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning.Advances in neural information processing systems, 36:34892–34916, 2023a. URLhttps://proceedings.neurips.cc/paper_ files/paper/2023...

    2023

  27. [2021]

    9 Vincent Cartillier, Zhile Ren, Neha Jain, Stefan Lee, Irfan Essa, and Dhruv Batra

    URL https://openaccess.thecvf.com/content/ICCV2021/html/Caron_Emerging_Properties_ in_Self-Supervised_Vision_Transformers_ICCV_2021_paper. 9 Vincent Cartillier, Zhile Ren, Neha Jain, Stefan Lee, Irfan Essa, and Dhruv Batra. Semantic MapNet: Building allocentric semantic maps a...

    2021 arXiv

  28. [2022]

    27 Abhijit Kundu, Yin Li, Frank Dellaert, Fuxin Li, and James M Rehg

    URL https://arxiv.org/abs/2211.15876. 27 Abhijit Kundu, Yin Li, Frank Dellaert, Fuxin Li, and James M Rehg. Joint semantic segmentation and 3d reconstruction from monocular video. InEuropean Conference on Computer Vision, pp. 703–718. Springer,

    arXiv

  29. [2023]

    28 43 Qiao Gu, Ali Kuwajerwala, Sacha Morin, Krishna Murthy Jatavallabhula, Bipasha Sen, Aditya Agarwal, Corban Rivera, William Paul, Kirsty Ellis, Rama Chellappa, et al

    URL https://openreview.net/pdf?id=b_CQDy9vrD1. 28 43 Qiao Gu, Ali Kuwajerwala, Sacha Morin, Krishna Murthy Jatavallabhula, Bipasha Sen, Aditya Agarwal, Corban Rivera, William Paul, Kirsty Ellis, Rama Chellappa, et al. Conceptgraphs: Open-vocabulary 3d scene graphs for percepti...

    2024

  30. [2024]

    23 Ze Gong and Yu Zhang

    URL https://arxiv.org/abs/2411.05755. 23 Ze Gong and Yu Zhang. Temporal spatial inverse semantics for robots communicating with humans. In IEEE International Conference on Robotics and Automation (ICRA), pp. 4451–4458. IEEE, 2018. URL https://ieeexplore.ieee.org/abstract/docum...

    2018 arXiv

  31. [2025]

    3, 16, 17, 23, 27, 29, 30 Allen Z Ren, Jaden Clark, Anushri Dixit, Masha Itkina, Anirudha Majumdar, and Dorsa Sadigh

    URL https://arxiv.org/abs/2411.07848. 3, 16, 17, 23, 27, 29, 30 Allen Z Ren, Jaden Clark, Anushri Dixit, Masha Itkina, Anirudha Majumdar, and Dorsa Sadigh. Explore until Confident: Efficient Exploration for Embodied Question Answering.arXiv preprint arXiv:2403.15941,

    arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.