Pith. sign in

REVIEW 3 major objections 5 minor 44 references

RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Base

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper claims that treating a CAD model as a retrievable multimodal knowledge base—visual appearance from rendered views plus 3D geometry—and fusing retrieved features into the decoder gives state-of-the-art 6D pose estimates, with…

desk verdict A genuine new mechanism with decent ablations; the SOTA claim outruns the evidence because every number is a single run. read the letter →

arxiv 2506.18856 v1 pith:RWVSENPK submitted 2025-06-23 cs.CV

Kuanning Wang , Yuqian Fu , Tianyu Wang , Yanwei Fu , Longfei Liang , Yu-Gang Jiang , Xiangyang Xue This is my paper · ORCID
classification cs.CV
keywords 6Dposeestimationretrieval-augmentedgenerationCADknowledgebasecross-attentionretrievalcontrastivelearningocclusionrobustnessroboticgrasping
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 tries to show that a 6D pose estimator works better when the object's CAD model is treated as an actively queried memory rather than only as a supervision signal. It builds a multimodal knowledge base by rendering the CAD from many viewpoints, extracting visual features with a frozen pretrained model, and attaching 3D coordinates and color to each point. A retrieval module then uses cross-attention to pull out the CAD features most relevant to the query image, and those retrieved features are fused into the pose-decoding pipeline. If the claim is right, the practical payoff is that occluded and textureless objects—exactly the failure cases for pose estimation—get a reliable source of extra visual and geometric context, and the reported results and robot-grasping tests support that.

What carries the argument

The ReSPC module carries the argument. Given the offline-built multimodal CAD knowledge base $F_b$ (per-point DINOv2 visual features from multi-view renderings plus 3D coordinates and color), it applies multi-head self-attention to capture global and local structure, enriches the result with a PointNet guided by global image appearance features, and then runs multi-head cross-attention $F_r=\operatorname{CrossAttn}(F_i,F_{pn},F_{pn})$ where the query image feature $F_i$ selects the CAD surface points relevant to the image. The retrieved $F_r$ is concatenated with $F_i$ and fed to the decoders, while the key features $F_k$ used for contrastive learning are built from the same knowledge base through Siren MLP layers, so CAD appearance and geometry are injected at both retrieval and decoding.

What would settle it

A concrete check: on LM-O, replace the retrieved features $F_r$ with features retrieved from an unrelated CAD model while keeping all other weights frozen, and measure Average Recall; if it does not fall to near the reported 66.5 of the no-retrieval ablation, the cross-attention retrieval is not the operative mechanism.

Watch

Extended reading notes

Core claim

The paper claims that a 6D pose estimator can be improved by actively retrieving visual and geometric information from the target object's CAD model instead of using CAD only for supervision. Its pipeline builds a multimodal knowledge base $F_b$: for each point of the CAD point cloud it stores a DINOv2 visual feature averaged over multi-view renderings, along with 3D coordinates and color. The ReSPC module enriches this base with self-attention and PointNet, then cross-attends from the query image feature $F_i$ to produce retrieved features $F_r=\operatorname{CrossAttn}(F_i,F_{pn},F_{pn})$, which are concatenated with $F_i$ and decoded. Trained with the SurfEmb-style contrastive loss and evaluated on five standard benchmarks, the method reports the highest average recall among the compared RGB methods on every dataset—70.0 on LM-O, 60.1 on IC-BIN, 83.3 on TUD-L, 68.6 on YCB-V, and 85.3 on HB—and the ablations show that removing the CAD knowledge base or replacing cross-attention fusion with simpler fusions lowers the score.

Load-bearing premise

The load-bearing premise is that DINOv2 features taken from synthetic CAD renderings and from real query photos live in a shared enough feature space that cross-attention can pick out the CAD surface points actually visible in the photo, even though that alignment is never directly supervised.

Editorial extensions

If this is right

  • On the five RGB benchmarks tested (LM-O, IC-BIN, TUD-L, YCB-V, HB), the method reports higher average recall than every compared prior method, making retrieval-augmented decoding the best-performing recipe among those compared.
  • The largest single-dataset gain over the SurfEmb baseline is on LM-O (70.0 vs 65.6 average recall), a heavily occluded set, so the visual retrieval appears to help most when only part of the object is visible.
  • With RGB-D input and ICP refinement, the same model reaches 76.8 on LM-O, 68.7 on IC-BIN, and ties the best reported 93.9 on TUD-L, showing the retrieval benefit persists after geometric refinement.
  • A single model with shared parameters serves all objects on a dataset; adding an object's CAD view renders builds its knowledge base without training an object-specific network, which would lower the cost of scaling to new objects.

Reading between the lines

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

  • The same retrieve-then-decode pattern could transfer to other tasks where a known 3D model accompanies a 2D observation, such as category-level pose estimation, object tracking, or robotic manipulation planning.
  • Because the knowledge-base features are built once offline from rendered views, the method's inference cost is concentrated in the retrieval cross-attention; a natural next question is how retrieval degrades with fewer rendered views or with CAD models that differ from the observed object.
  • The reported per-dataset gains are uneven, so a profitable test is to stratify results by occlusion level to see whether retrieval's benefit is concentrated in the most occluded frames, which the paper does not report.
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. The paper proposes RAG-6DPose, a monocular 6D pose estimation method that treats the 3D CAD model as a retrievable knowledge base. It first builds an offline multi-modal CAD feature bank by rendering the CAD model from multiple views, extracting DINOv2 visual features, and reprojecting them onto 3D points together with positions and colors. At inference, a ReSPC module (self-attention, PointNet, and cross-attention) retrieves relevant CAD features conditioned on the query image, and the retrieved features are fused into a SurfEmb-style decoder for dense correspondence prediction. Training uses a contrastive loss between decoded query features and CAD key features. The paper reports Average Recall on LM-O, IC-BIN, TUD-L, HB, and YCB-V, ablation studies on LM-O, and real-robot grasping trials, claiming state-of-the-art results across all five datasets.

Significance. If the reported numbers are representative, the paper makes a useful empirical contribution: it demonstrates that combining DINOv2 visual features of CAD renderings with geometric point features can improve dense-correspondence pose estimation under occlusion. The ablation study is a real strength, especially the '−3D CAD Features' row and the 'C.A. Fusion→Avg' control, which together suggest that the retrieval module, not just added parameters, drives the gain. The breadth of BOP datasets and the inclusion of real-robot trials are also positive. The main weakness is that every numerical claim rests on single runs without variance or code release, so the central 'state-of-the-art' conclusion is not yet statistically established; the specific worry about synthetic-to-real domain gap in DINOv2 features is not directly resolved by the paper's diagnostics.

major comments (3)
  1. [IV-B, Tables I and II] All reported Average Recall values are single-run numbers without standard deviations, confidence intervals, or seed counts, and no code or trained models are released. The margins over the closest competitor are small on several datasets (YCB-V: 68.6 vs 68.1; IC-BIN: 60.1 vs 58.5; TUD-L: 83.3 vs 82.3). In training stochastic 6D pose estimators, run-to-run variation of this size is common, so the statement that the method 'outperforms all the competitors across five datasets' is not statistically supported. The authors should provide multi-seed results with error bars (at least three to five seeds) and ideally release code or models; without this, the central empirical claim is not verifiable.
  2. [III-C and IV-D] The paper describes ReSPC as retrieving the CAD surface points visible in the query, but no direct evidence of retrieval quality is presented. There are no attention visualizations, no retrieval-precision metrics, and no comparison against an oracle retriever, so a reader cannot tell whether the cross-attention weights select the geometrically correct CAD points or simply act as a learned feature-fusion mechanism. The control 'C.A. Fusion→Avg' (Table IV) removes the cross-attention and loses 2.3 AR, but that ablation changes the fusion operation and parameter count simultaneously; it does not measure retrieval accuracy. Because the synthetic-to-real domain gap in DINOv2 features is the main correctness risk of the design, the authors should add direct retrieval diagnostics (e.g., attention maps over CAD points versus ground-truth visible surface, or retrieval precision) to support the mechanistic claim.
  3. [IV-A and IV-D] The training description is underspecified: 'trained to convergence' with no number of iterations, batch size, augmentation, learning-rate schedule, or validation criterion, and the full-scale ablation says 'we conduct 5 experiments' while Table IV lists six rows. Combined with the absence of code, this makes the experiments difficult to reproduce and prevents independent verification of the ablations. Please supply complete training details or release the code.
minor comments (5)
  1. [Table I caption] The caption reads 'P.E. MEANS THE NUMBER OF POSE ESTIMATORS FOR ANN-OBJECTS DATASET', which is unclear and contains a likely typo ('ANN-OBJECTS' should probably be 'AN OBJECTS' or 'ANNOTATED OBJECTS'); please clarify what the P.E. column counts.
  2. [IV-B and Figure 5] There are several typos: 'DPOPv2' in the subsection 'Comparison with 2D-3D Correspondence Methods' should be 'DPODv2', and the figure label 'KINOV A GEN2' should be 'KINOVA GEN2'. The rendered equation in Figure 1 also appears garbled.
  3. [III-D] In the definition of key features, the Siren layers S_g, S_v, and S_i are introduced without stating their input and output dimensions; please specify these dimensions for reproducibility.
  4. [Table III] Table III shows that RAG-6DPose is not the best on every object at either threshold (e.g., eggbox and helicopter at θ=10), so the text saying 'best results on most objects' is accurate, but the aggregate phrase 'outperforms all competitors' should be qualified to avoid overstatement.
  5. [IV-A and IV-D] The paper does not report inference time, parameter count, or the number of attention heads and feature dimensions in ReSPC; adding these details would help assess the method's practicality and reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the SOTA claim is benchmark-measured against external baselines, and the retrieval features are trained via ground-truth correspondences, not defined in terms of the final pose.

full rationale

RAG-6DPose is an empirical systems paper; its central claim is that retrieving CAD visual and geometric features via the ReSPC module improves 6D pose estimation. This is supported by BOP Average Recall comparisons against external baselines (Tables I and II) and by ablations (Table IV). No equation defines the retrieved features Fr as the pose output: Fr = CrossAttn(Fi, Fpn, Fpn) is trained indirectly through the InfoNCE contrastive loss Lcon using ground-truth visible object coordinates as positives, and inference uses PnP-RANSAC on sampled correspondences. Thus the retrieval is not circularly defined by the final pose. The use of DINOv2 both for query encoding and for knowledge-base construction is a shared-representation design choice, not a circular step, since nothing in the loss forces the retrieval to match the pose by construction; it only encourages feature consistency with ground-truth 2D-3D correspondences. The only author-overlapping citation is Domain-RAG [19] in the Related Work section, used to illustrate retrieval-augmented image generation; it is not load-bearing for any experimental or architectural claim. The absence of variance statistics and code is a legitimate concern about the strength of the empirical SOTA claim, but that is a statistical robustness issue, not circularity. No self-definitional reduction, fitted-input-called-prediction step, imported uniqueness theorem, or ansatz-smuggling citation can be exhibited from the paper's text.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claim rests on the SurfEmb backbone, on the transferability of DINOv2 features across domains, and on the correctness of the nearest-neighbor feature reprojection. These are reasonable engineering assumptions, but they are taken as given rather than validated independently.

free parameters (2)
  • Alpha (loss weighting factor) = not reported
    Balances segmentation loss and contrastive loss in L = L_con + alpha * L_m (Section III-D). Chosen by hand; no value or sensitivity analysis given.
  • Number of rendered views m and point count N_p = not reported
    Knowledge-base construction uses m multi-view renderings and N_p CAD points; these densities affect retrieval quality and are not specified in the paper.
assumptions (3)
  • domain assumption SurfEmb's 2D-3D contrastive correspondence framework is a valid and strong backbone for pose estimation.
    Section III-A states 'we establish our method following a similar pipeline with SurfEmb' and inherits its contrastive learning; any weakness in SurfEmb's formulation is inherited.
  • domain assumption DINOv2 features extracted from synthetic CAD renders and real query images live in a shared representation space suitable for cross-attention retrieval.
    Section III-C uses the same frozen DINOv2 for the knowledge base and the query image; the paper does not finetune DINOv2 and does not provide evidence of cross-domain alignment beyond downstream task performance.
  • domain assumption The nearest-neighbor reprojection from rendered depth point clouds to CAD points assigns each CAD point the correct multi-view visual feature.
    Section III-B assigns the feature of the nearest rendered-depth point to each CAD point; if the renderings are sparse or misaligned, some CAD points receive averaged or incorrect features.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Base." pith.science (2026). https://pith.science/paper/RWVSENPK

@misc{pith2026250618856,
  author       = {Pith},
  title        = {Pith review of: RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Base},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RWVSENPK}},
  note         = {Machine review of arXiv:2506.18856}
}
read the original abstract

Accurate 6D pose estimation is key for robotic manipulation, enabling precise object localization for tasks like grasping. We present RAG-6DPose, a retrieval-augmented approach that leverages 3D CAD models as a knowledge base by integrating both visual and geometric cues. Our RAG-6DPose roughly contains three stages: 1) Building a Multi-Modal CAD Knowledge Base by extracting 2D visual features from multi-view CAD rendered images and also attaching 3D points; 2) Retrieving relevant CAD features from the knowledge base based on the current query image via our ReSPC module; and 3) Incorporating retrieved CAD information to refine pose predictions via retrieval-augmented decoding. Experimental results on standard benchmarks and real-world robotic tasks demonstrate the effectiveness and robustness of our approach, particularly in handling occlusions and novel viewpoints. Supplementary material is available on our project website: https://sressers.github.io/RAG-6DPose .

Figures

Figures reproduced from arXiv: 2506.18856 by the authors.

Figure 1
Figure 1. Comparison of methods: Prior works (a) and (b) use CAD for [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Method Architecture. The left part provides an overview of “Building a Multi-Modal CAD Knowledge Base.” Particularly, we have two processes, “Retrieving CAD Information” and “Incorporating Retrieved CAD for Pose Estimation” to leverage the knowledge base Fb. The diagonal dashed line separates the two processes, highlighting their distinct functions in our method. B. Offline Multi-Modal Knowledge Base Construction To… view at source ↗
Figure 3
Figure 3. ReSPC Module Architecture. This module takes the CAD knowledge base Fb and image features as input and outputs the retrieved features Fr. The left part illustrates how we use encoders to extract image features. The middle part demonstrates how attention mechanisms and PointNet process Fb. The right part details the function of Cross-Attention, illustrating the core retrieval process from the feature Fpn extracted fr… view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Robotic Experimental Setup. Left: a third-person view of the setup. A first-person camera captures RGB image observations. We show different 3D printed objects from the LM-O and YCB-V datasets and a pen, all of which are used in our experiments. Right: we show “Level 2…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 42 canonical work pages

  1. [1]

    Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes,

    Y . Xiang, T. Schmidt, V . Narayanan, and D. Fox, “Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes,”ArXiv, 2017

    work page 2017

  2. [2]

    Mrc-net: 6-dof pose estimation with multiscale residual correlation,

    Y . Li, Y . Mao, R. Bala, and S. Hadap, “Mrc-net: 6-dof pose estimation with multiscale residual correlation,”ArXiv, 2024

    work page 2024

  3. [3]

    Cdpn: Coordinates-based disentangled pose network for real-time rgb-based 6-dof object pose estimation,

    Z. Li, G. Wang, and X. Ji, “Cdpn: Coordinates-based disentangled pose network for real-time rgb-based 6-dof object pose estimation,”2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019

    work page 2019

  4. [4]

    Dcl-net: Deep correspondence learning network for 6d pose estimation,

    H. Li, J. Lin, and K. Jia, “Dcl-net: Deep correspondence learning network for 6d pose estimation,” inEuropean Conference on Computer Vision, 2022

    work page 2022

  5. [5]

    Center-based decoupled point cloud registration for 6d object pose estimation,

    H. Jiang, Z. Dang, S. Gu, J. Xie, M. Salzmann, and J. Yang, “Center-based decoupled point cloud registration for 6d object pose estimation,”2023 IEEE/CVF International Conference on Computer Vision (ICCV), 2023

    work page 2023

  6. [6]

    Dinov2: Learning robust visual features without supervision,

    M. Oquab, T. Darcet, T. Moutakanni, H. Q. V o, M. Szafraniec, V . Khalidov, P. Fernandez, D. Haziza, F. Massa, A. El-Nouby, M. As- sran, N. Ballas, W. Galuba, R. Howes, P.-Y . B. Huang, S.-W. Li, I. Misra, M. G. Rabbat, V . Sharma, G. Synnaeve, H. Xu, H. J ´egou, J. Mairal, P. Labatut, A. Joulin, and P. Bojanowski, “Dinov2: Learning robust visual features...

    work page 2023

  7. [7]

    Posenet: A convolutional network for real-time 6-dof camera relocalization,

    A. Kendall, M. K. Grimes, and R. Cipolla, “Posenet: A convolutional network for real-time 6-dof camera relocalization,”2015 IEEE Inter- national Conference on Computer Vision (ICCV), 2015

    work page 2015

  8. [8]

    Gdr-net: Geometry- guided direct regression network for monocular 6d object pose esti- mation,

    G. Wang, F. Manhardt, F. Tombari, and X. Ji, “Gdr-net: Geometry- guided direct regression network for monocular 6d object pose esti- mation,”2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021

    work page 2021

Show all 44 references
  1. [9]

    Pvnet: Pixel-wise voting network for 6dof pose estimation,

    S. Peng, Y . Liu, Q.-X. Huang, H. Bao, and X. Zhou, “Pvnet: Pixel-wise voting network for 6dof pose estimation,”2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2018

    2019

  2. [10]

    Pix2pose: Pixel-wise coordinate regression of objects for 6d pose estimation,

    K. Park, T. Patten, and M. Vincze, “Pix2pose: Pixel-wise coordinate regression of objects for 6d pose estimation,”2019 IEEE/CVF Inter- national Conference on Computer Vision (ICCV), 2019

    2019

  3. [11]

    Surfemb: Dense and continuous correspondence distributions for object pose estimation with learnt surface embeddings,

    R. L. Haugaard and A. G. Buch, “Surfemb: Dense and continuous correspondence distributions for object pose estimation with learnt surface embeddings,”2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021

    2022

  4. [12]

    Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,

    M. A. Fischler and R. C. Bolles, “Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography,”Communications of the ACM, 1981

    1981

  5. [13]

    Dpodv2: Dense correspondence-based 6 dof pose estimation,

    I. S. Shugurov, S. Zakharov, and S. Ilic, “Dpodv2: Dense correspondence-based 6 dof pose estimation,”IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021

    2021

  6. [14]

    Prior-information- guided corresponding point regression network for 6d pose estima- tion,

    H. Gan, L. Wang, Y . Su, W. Ruan, and X. Jiao, “Prior-information- guided corresponding point regression network for 6d pose estima- tion,”Comput. Graph., 2024

    2024

  7. [15]

    Retrieval-augmented generation for knowledge-intensive nlp tasks,

    P. Lewis, E. Perez, A. Piktus, F. Petroni, V . Karpukhin, N. Goyal, H. Kuttler, M. Lewis, W. tau Yih, T. Rockt ¨aschel, S. Riedel, and D. Kiela, “Retrieval-augmented generation for knowledge-intensive nlp tasks,”ArXiv, 2020

    2020

  8. [16]

    Retrieval augmented generation and understanding in vision: A survey and new outlook,

    X. Zheng, Z. Weng, Y . Lyu, L. Jiang, H. Xue, B. Ren, D. Paudel, N. Sebe, L. Van Gool, and X. Hu, “Retrieval augmented generation and understanding in vision: A survey and new outlook,”arXiv preprint, 2025

    2025

  9. [17]

    Ra-clip: Retrieval augmented contrastive language-image pre-training,

    C.-W. Xie, S. Sun, X. Xiong, Y . Zheng, D. Zhao, and J. Zhou, “Ra-clip: Retrieval augmented contrastive language-image pre-training,”2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    2023

  10. [18]

    Realrag: Retrieval-augmented realistic image generation via self-reflective contrastive learning,

    Y . Lyu, X. Zheng, L. Jiang, Y . Yan, X. Zou, H. Zhou, L. Zhang, and X. Hu, “Realrag: Retrieval-augmented realistic image generation via self-reflective contrastive learning,”arXiv preprint, 2025

    2025

  11. [19]

    Domain-rag: Retrieval-guided compositional image generation for cross-domain few-shot object detection,

    Y . Li, X. Qiu, Y . Fu, J. Chen, T. Qian, X. Zheng, D. P. Paudel, Y . Fu, X. Huang, L. Van Goolet al., “Domain-rag: Retrieval-guided compositional image generation for cross-domain few-shot object detection,”arXiv preprint, 2025

    2025

  12. [20]

    Beyond text: Optimizing rag with multi- modal inputs for industrial applications,

    M. Riedler and S. Langer, “Beyond text: Optimizing rag with multi- modal inputs for industrial applications,” 2024

    2024

  13. [21]

    Implicit neural representations with periodic activation functions,

    V . Sitzmann, J. N. P. Martel, A. W. Bergman, D. B. Lindell, and G. Wetzstein, “Implicit neural representations with periodic activation functions,”ArXiv, 2020

    2020

  14. [22]

    D3fields: Dynamic 3d descriptor fields for zero-shot generalizable robotic manipulation,

    Y . Wang, Z. Li, M. Zhang, K. R. Driggs-Campbell, J. Wu, F.-F. Li, and Y . Li, “D3fields: Dynamic 3d descriptor fields for zero-shot generalizable robotic manipulation,”ArXiv, vol. abs/2309.16118, 2023

    2023 arXiv

  15. [23]

    Freeze: Training- free zero-shot 6d pose estimation with geometric and vision foundation models,

    A. Caraffa, D. Boscaini, A. Hamza, and F. Poiesi, “Freeze: Training- free zero-shot 6d pose estimation with geometric and vision foundation models,” 2023

    2023

  16. [24]

    V oxposer: Composable 3d value maps for robotic manipulation with language models,

    W. Huang, C. Wang, R. Zhang, Y . Li, J. Wu, and L. Fei-Fei, “V oxposer: Composable 3d value maps for robotic manipulation with language models,”ArXiv, 2023

    2023

  17. [25]

    Pointnet: Deep learning on point sets for 3d classification and segmentation,

    C. Qi, H. Su, K. Mo, and L. J. Guibas, “Pointnet: Deep learning on point sets for 3d classification and segmentation,”2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016

    2017

  18. [26]

    Aggregated resid- ual transformations for deep neural networks,

    S. Xie, R. B. Girshick, P. Doll ´ar, Z. Tu, and K. He, “Aggregated resid- ual transformations for deep neural networks,”2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016

    2017

  19. [27]

    U-net: Convolutional networks for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,”ArXiv, 2015

    2015

  20. [28]

    Representation learning with contrastive predictive coding,

    A. van den Oord, Y . Li, and O. Vinyals, “Representation learning with contrastive predictive coding,”ArXiv, 2018

    2018

  21. [29]

    Ep n p: An accurate o (n) solution to the p n p problem,

    V . Lepetit, F. Moreno-Noguer, and P. Fua, “Ep n p: An accurate o (n) solution to the p n p problem,”International journal of computer vision, 2009

    2009

  22. [30]

    Learning 6d object pose estimation using 3d object coordinates,

    E. Brachmann, A. Krull, F. Michel, S. Gumhold, J. Shotton, and C. Rother, “Learning 6d object pose estimation using 3d object coordinates,” inEuropean Conference on Computer Vision, 2014

    2014

  23. [31]

    Recovering 6d object pose and predicting next-best-view in the crowd,

    A. Doumanoglou, R. Kouskouridas, S. Malassiotis, and T.-K. Kim, “Recovering 6d object pose and predicting next-best-view in the crowd,”2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015

    2016

  24. [32]

    Homebreweddb: Rgb-d dataset for 6d pose estimation of 3d objects,

    R. Kaskman, S. Zakharov, I. S. Shugurov, and S. Ilic, “Homebreweddb: Rgb-d dataset for 6d pose estimation of 3d objects,”2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), 2019

    2019

  25. [33]

    Bop: Benchmark for 6d object pose estimation,

    T. Hodan, F. Michel, E. Brachmann, W. Kehl, A. G. Buch, D. Kraft, B. Drost, J. Vidal, S. Ihrke, X. Zabulis, C. Sahin, F. Manhardt, F. Tombari, T.-K. Kim, J. Matas, and C. Rother, “Bop: Benchmark for 6d object pose estimation,”ArXiv, 2018

    2018

  26. [34]

    Bop challenge 2023 on detection, segmentation and pose estimation of seen and unseen rigid objects,

    T. Hodan, M. Sundermeyer, Y . Labbe, V . N. Nguyen, G. Wang, E. Brachmann, B. Drost, V . Lepetit, C. Rother, and J. Matas, “Bop challenge 2023 on detection, segmentation and pose estimation of seen and unseen rigid objects,” 2024

    2023

  27. [35]

    Cnos: A strong baseline for cad-based novel object segmentation,

    V . N. Nguyen, T. Hodan, G. Ponimatkin, T. Groueix, and V . Lepetit, “Cnos: A strong baseline for cad-based novel object segmentation,” 2023 IEEE/CVF International Conference on Computer Vision Work- shops (ICCVW), 2023

    2023

  28. [36]

    Bop challenge 2020 on 6d object localization,

    T. Hodan, M. Sundermeyer, B. Drost, Y . Labb ´e, E. Brachmann, F. Michel, C. Rother, and J. Matas, “Bop challenge 2020 on 6d object localization,”ArXiv, 2020

    2020

  29. [37]

    Neural correspondence field for object pose estimation,

    L. Huang, T. Hodan, L. Ma, L. Zhang, L. Tran, C. D. Twigg, P.-C. Wu, J. Yuan, C. Keskin, and R. Wang, “Neural correspondence field for object pose estimation,” inEuropean Conference on Computer Vision, 2022

    2022

  30. [38]

    Cosypose: Consistent multi-view multi-object 6d pose estimation,

    Y . Labb’e, J. Carpentier, M. Aubry, and J. Sivic, “Cosypose: Consistent multi-view multi-object 6d pose estimation,” inEuropean Conference on Computer Vision, 2020

    2020

  31. [39]

    So- pose: Exploiting self-occlusion for direct 6d pose estimation,

    Y . Di, F. Manhardt, G. Wang, X. Ji, N. Navab, and F. Tombari, “So- pose: Exploiting self-occlusion for direct 6d pose estimation,”2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021

    2021

  32. [40]

    Crt-6d: Fast 6d object pose estimation with cascaded refinement transformers,

    P. Castro and T.-K. Kim, “Crt-6d: Fast 6d object pose estimation with cascaded refinement transformers,”2023 IEEE/CVF Winter Confer- ence on Applications of Computer Vision (WACV), 2022

    2023

  33. [41]

    Perspective flow aggregation for data-limited 6d object pose estimation,

    Y . Hu, P. Fua, and M. Salzmann, “Perspective flow aggregation for data-limited 6d object pose estimation,” inEuropean Conference on Computer Vision, 2022

    2022

  34. [42]

    Yolo-6d-pose: Enhancing yolo for single-stage monocular multi-object 6d pose estimation,

    D. Maji, S. Nagori, M. Mathew, and D. Poddar, “Yolo-6d-pose: Enhancing yolo for single-stage monocular multi-object 6d pose estimation,”2024 International Conference on 3D Vision (3DV), 2024

    2024

  35. [43]

    Efficient variants of the icp algo- rithm,

    S. Rusinkiewicz and M. Levoy, “Efficient variants of the icp algo- rithm,”Proceedings Third International Conference on 3-D Digital Imaging and Modeling, 2001

    2001

  36. [44]

    Corri2p: Deep image-to-point cloud registration via dense correspondence,

    S. Ren, Y . Zeng, J. Hou, and X. Chen, “Corri2p: Deep image-to-point cloud registration via dense correspondence,”IEEE Transactions on Circuits and Systems for Video Technology, 2022

    2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.