Pith. sign in

REVIEW 3 major objections 5 minor 3 cited by

DeOcc-1-to-3: 3D De-Occlusion from a Single Image via Self-Supervised Multi-View Diffusion

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 ↗

Pith's one-line read DeOcc-1-to-3 turns one occluded photo into six consistent 3D-ready views, without inpainting or labels.

desk verdict A plausible self-supervised de-occlusion method with a useful new benchmark, but the evaluation lacks the controls needed to back the central claim; worth sending to review with a request for major revision. read the letter →

arxiv 2506.21544 v2 pith:TBP7QLZ5 submitted 2025-06-26 cs.CV

Yansong Qu , Shaohui Dai , Xinyang Li , Yuze Wang , You Shen , Liujuan Cao , Rongrong Ji This is my paper
classification cs.CV
keywords 3Dreconstructionde-occlusionmulti-viewdiffusionself-supervisedlearningocclusion-awareviewsynthesissingle-imagemodels
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

DeOcc-1-to-3 claims that a single partially occluded photograph is enough to generate six mutually consistent de-occluded views of the object, and that those views support reliable 3D reconstruction. The paper argues that the usual two-stage recipe—first inpaint the occluded area in 2D, then run view synthesis—accumulates errors and lacks 3D awareness. Instead, it fine-tunes an existing multi-view diffusion model end-to-end on occluded–clean image pairs, with a frozen teacher producing six-view pseudo-ground truths from the clean images. On the new Occ-LVIS benchmark, the method outperforms both the vanilla image-to-3D pipeline and the two-stage Pix2Gestalt pipeline on all reported 2D and 3D metrics, at no extra inference cost.

What carries the argument

The load-bearing mechanism is teacher–student self-distillation through a frozen multi-view diffusion model G (Zero123++). G converts clean images Ifull into six-view pseudo-ground-truths G(Ifull); the student, initialized from the same weights, is fully fine-tuned to produce the same six views from occluded inputs Iocc, using a velocity-based denoising objective. The six predefined camera poses (elevations {30°, −20°}, azimuths every 60°) anchor cross-view consistency across all outputs. The training data is built with SAM-segmented objects composited onto backgrounds with random occlusion masks, and the paper filters samples where the foreground is inherently incomplete or touches the image boundary.

What would settle it

Train the identical student architecture and fine-tuning schedule using ground-truth six-view renderings of synthetic 3D objects (with synthetic occlusions) instead of the teacher's pseudo-ground truths, then evaluate on Occ-LVIS. If the gains over the vanilla Zero123++ baseline disappear or reverse, the pseudo-label teacher is what carries the reported improvement; if the gains persist, the method's occlusion reasoning is independent of the teacher's specific outputs.

Watch

Extended reading notes

Core claim

The central claim is that a multi-view diffusion model can be made occlusion-aware without any architectural change by fully fine-tuning it to map a single occluded RGB image directly to six predefined-posed novel views that are both completed and view-consistent. The training signal is self-supervised: for each clean image, a frozen teacher view-synthesis model G produces pseudo-ground-truth views G(Ifull); random occlusions are applied to the clean image to obtain Iocc; and the student is trained with the standard denoising objective to output G(Ifull) from Iocc. The paper reports that this single-stage design beats the two-stage 2D-inpainting-plus-view-synthesis baseline on Occ-LVIS, improving FID from 30.19 to 29.08, CLIP from 0.7833 to 0.7892, and V-IoU from 0.3232 to 0.3445, while keeping the same runtime as the no-occlusion-handling baseline.

Load-bearing premise

The entire supervision signal is defined by the frozen teacher model G: the student is trained to match G's six-view outputs on clean images, so if G systematically hallucinates or fails on a shape, the student learns to reproduce that failure rather than true occlusion-completed geometry.

Editorial extensions

If this is right

  • Occlusion handling can be folded directly into the view synthesis stage, eliminating the need for separate 2D inpainting and avoiding its error accumulation.
  • Any downstream reconstruction backend that consumes multi-view images—InstantMesh, NeRF, or 3D Gaussian Splatting—can accept DeOcc-1-to-3 outputs without modification.
  • The self-supervised training recipe requires no manual labels, only occluded–clean image pairs and a pretrained multi-view generator.
  • On the Occ-LVIS benchmark, the method improves all reported 2D and 3D metrics over both baselines, with no increase in inference time or parameters relative to the vanilla pipeline.

Reading between the lines

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

  • If the teacher model's pseudo-ground truths are biased—for example on unusual object shapes—those biases are inherited directly by the student; the reported gains may partly reflect self-distillation rather than newly learned occlusion reasoning.
  • The recipe is architecture-agnostic: swapping Zero123++ for a stronger multi-view diffusion backbone could scale the benefits, provided the teacher's pseudo-labels improve correspondingly.
  • A natural stress test is to train with ground-truth multi-view renderings from synthetic 3D assets instead of teacher pseudo-labels, which would isolate how much of the improvement is due to occlusion-aware fine-tuning and how much to the teacher's prior.
  • Because the benchmark occludes rendered objects with other foreground objects, real-world generalization may be lower; testing on natural occluded photographs at matched occlusion ratios would separate benchmark-specific gains from genuine robustness.
Share X Bluesky LinkedIn Reddit HN

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 DeOcc-1-to-3, a method for 3D de-occlusion from a single image. It fine-tunes Zero123++ so that the model directly synthesizes six multi-view-consistent, de-occluded images from an occluded input, which are then fed to InstantMesh for 3D reconstruction. Training is self-supervised: occluded/clean pairs are created by overlaying SAM-segmented foreground objects onto backgrounds, and a frozen teacher multi-view diffusion model G generates pseudo-ground-truth views G(Ifull) from the clean image; a student model with the same architecture is trained with a denoising loss conditioned on the occluded image. The paper also introduces Occ-LVIS, a benchmark with five occlusion levels, and reports consistent improvements over Zero123+++InstantMesh and Pix2Gestalt+Zero123+++InstantMesh on 2D metrics (CLIP, FID, KID) and 3D metrics (CD, F-Score, V-IoU).

Significance. DeOcc-1-to-3 is a clean and practical idea: instead of decoupling 2D inpainting from view synthesis, it jointly learns completion and multi-view generation in a single diffusion model, requiring no manual annotations and no architectural changes. If the reported gains are robust, the approach is a useful recipe for occlusion-aware view synthesis and can be plugged into existing reconstruction pipelines, with an efficiency advantage over two-stage baselines. The introduction of a standardized benchmark with stratified occlusion levels is also a valuable community resource. However, the evaluation does not currently establish the central causal claim that the improvements come from a learned occlusion-completion mechanism rather than from self-distillation of the teacher's clean-image prior; the paper offers no code, no error bars, and no ablations of its key design choices.

major comments (3)
  1. [Sec. 3.3 and Sec. 3.4, Eq. (1)] The entire supervision signal is defined by the frozen teacher G: pseudo-ground-truth views are G(Ifull), and the student is initialized from G itself. The paper never measures the teacher's own accuracy on clean inputs (e.g., D(G(Ifull), GT renders) on Occ-LVIS), never compares against a student trained with true ground-truth multi-view renders, and never reports the oracle upper bound of running G on the clean image. Since the student's ceiling is exactly G's clean-image quality, the improvements in Tables 2-3 could reflect the student inheriting the teacher's prior rather than a genuine occlusion-completion mechanism. The Limitations section concedes suboptimal behavior on unusual shapes but does not quantify this label bias. Please add (i) teacher-versus-ground-truth fidelity numbers, (ii) a GT-supervised student ablation, and (iii) a clean-input oracle comparison; without these, the central claim that the model directly predicts reliable de-occluded views is not established.
  2. [Tables 2 and 3] All headline numbers are single-run point estimates with no error bars, no number of seeds, and no significance tests. The benchmark explicitly defines five occlusion levels (Table 1), yet no per-level breakdown is reported; with only three methods and three metrics, the improvements (e.g., FID 30.19 to 29.08, KID 0.0043 to 0.0035) are small enough that seed-level variance could change the ranking. Please report mean and standard deviation over at least three seeds and provide per-occlusion-level results. In addition, the two data-construction choices most likely to affect behavior—the inclusion of identity pairs (Sec. 3.3) and the dilation/erosion augmentation—are not ablated, so it is unclear which component drives the gain.
  3. [Sec. 3.6 and Sec. 4.1] Occ-LVIS is built from Objaverse-LVIS, and the teacher model G (Zero123++) was trained on Objaverse-derived data; evaluating on this distribution can inflate apparent gains because the pseudo-ground-truth is close to ground truth exactly where G is strong. The paper presents only qualitative evidence on real-world images (Figures 3 and 5). Please add a quantitative evaluation on a held-out distribution (e.g., real-world occluded photographs with estimated annotations, or a different 3D dataset) and discuss the distribution overlap between the pseudo-GT training set and the benchmark. This is important for the claim of generalization to real-world occlusions.
minor comments (5)
  1. [Sec. 3.1, Eq. (1)] The text describes the objective as 'velocity-based,' but Eq. (1) is the standard noise-prediction loss ||epsilon - epsilon_theta||^2; please correct the terminology or change the equation to a v-prediction form.
  2. [Sec. 3.6 and Table 1] The text defines occlusion ratio as 'the proportion of the target object’s visible area,' but the table's ranges (0–10%, 10–20%, ..., >=40%) read naturally as occluded-area proportions; please clarify the direction and naming to avoid misreading the benchmark's difficulty ordering.
  3. [Sec. 3.3] The text says 'we construct ... based on the SA-1B dataset [45],' but reference [45] is the Segment Anything Model paper; please cite the SA-1B dataset explicitly or state that SA-1B is introduced in [45].
  4. [Figure 2] The caption contains an apparent typo: 'Occluded ImageVisibleMask' should be 'Occluded Image, Visible Mask' or similar.
  5. [Sec. 4.1] For the 2D metrics (FID, KID, CLIP), please state explicitly what the reference distribution is (the ground-truth six views, the clean rendered views, or something else) and whether scores are computed per-view or on the tiled 3x2 images; this affects how the numbers should be interpreted.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central claim is evaluated against externally rendered ground-truth views, not against the teacher's pseudo-labels.

full rationale

The paper's training objective uses pseudo-ground-truth views G(Ifull) generated by a frozen teacher model G (Sec. 3.3), and the student is fine-tuned from the same pretrained weights to predict those views from an occluded input (Sec. 3.4, Eq. 1). This is a self-distillation setup, but it is not circular by construction: the paper explicitly calls G(Ifull) 'pseudo-ground-truth,' and the reported quantitative results compare the generated six views and downstream meshes against actual rendered views and geometry from the Occ-LVIS benchmark, which are independent of the teacher's outputs. The evaluation therefore does not reduce to the training labels. The teacher's accuracy does impose an upper bound on the student's clean-image fidelity, and the paper does not report D(G(Ifull), GT-render), but that is a correctness/risk concern about label bias, not a circularity of the kind where an equation or fitted parameter is renamed as a prediction. Self-citations in Related Works (e.g., SynergyAmodal) are not load-bearing for the main claim, and no uniqueness theorem or ansatz is imported from prior work to force the method's choice. The limitations section openly concedes degraded performance under extreme occlusion and unusual shapes, which is consistent with a genuinely empirical evaluation against external ground truth. Overall, the central derivation chain is self-contained and externally falsifiable, so the circularity score is low.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim is empirical and rests on several domain assumptions rather than a closed-form derivation. The main unverified premise is that the teacher model provides accurate supervision and that synthetic occlusions transfer to real scenes.

free parameters (5)
  • learning_rate = 2e-5
    AdamW optimizer setting; chosen by hand, not fitted to data.
  • batch_size = 32
    Training batch size; hand-chosen.
  • training_steps = 150k
    Total fine-tuning steps; hand-chosen.
  • ema_decay = 0.9999
    Exponential moving average decay; hand-chosen.
  • occlusion_level_thresholds = 0.1, 0.2, 0.3, 0.4
    Stratification boundaries for the Occ-LVIS benchmark (Table 1); hand-chosen.
assumptions (4)
  • domain assumption The frozen teacher G produces accurate six-view pseudo-ground-truths from clean images.
    Sec 3.3: training pairs are formed as <I_occ, G(I_full)>; the student's target is entirely defined by G's output. No error analysis of G is provided.
  • domain assumption Randomly composited SA-1B occlusions are representative of real-world occlusions.
    Sec 3.3: training data is built by overlaying SAM segments; generalization to natural occlusions (Fig. 5) is assumed and only qualitatively shown.
  • domain assumption Six predefined views (Zero123++ pose set) suffice for downstream reconstruction.
    Sec 3.1: output is a 3x2 tiled image at fixed elevations and azimuths; no view coverage analysis is given.
  • domain assumption Objaverse-LVIS renderings with random occluders are a valid benchmark for occlusion-aware 3D reconstruction.
    Sec 3.6: the benchmark is created in-house; no external validation or user study.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DeOcc-1-to-3: 3D De-Occlusion from a Single Image via Self-Supervised Multi-View Diffusion." pith.science (2026). https://pith.science/paper/TBP7QLZ5

@misc{pith2026250621544,
  author       = {Pith},
  title        = {Pith review of: DeOcc-1-to-3: 3D De-Occlusion from a Single Image via Self-Supervised Multi-View Diffusion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TBP7QLZ5}},
  note         = {Machine review of arXiv:2506.21544}
}
read the original abstract

Reconstructing 3D objects from a single image remains challenging, especially under real-world occlusions. While recent diffusion-based view synthesis models can generate consistent novel views from a single RGB image, they typically assume fully visible inputs and fail when parts of the object are occluded, resulting in degraded 3D reconstruction quality. We propose DeOcc-1-to-3, an end-to-end framework for occlusion-aware multi-view generation that synthesizes six structurally consistent novel views directly from a single occluded image, enabling reliable 3D reconstruction without prior inpainting or manual annotations. Our self-supervised training pipeline leverages occluded-unoccluded image pairs and pseudo-ground-truth views to teach the model structure-aware completion and view consistency. Without modifying the original architecture, we fully fine-tune the view synthesis model to jointly learn completion and multi-view generation. Additionally, we introduce the first benchmark for occlusion-aware reconstruction, covering diverse occlusion levels, object categories, and masking patterns, providing a standardized protocol for future evaluation.

Figures

Figures reproduced from arXiv: 2506.21544 by the authors.

Figure 1
Figure 1. DeOcc-1-to-3 takes a single occluded image (left) as input and synthesizes structurally consistent multi-view de-occluded images (middle). These outputs can be seamlessly integrated into various 3D reconstruction or generation frameworks to produce accurate meshes and surface normals (right). The proposed pipeline demonstrates generalization across diverse object categories and occlusion scenarios. Abstract Reconstr… view at source ↗
Figure 2
Figure 2. Overview of DeOcc-1-to-3. Top: Occluded images Iocc are generated by applying random occlusions to full images Ifull. A frozen multi-view diffusion model produces six-view pseudo￾ground-truths G(Ifull), forming training pairs⟨Iocc, G(Ifull)⟩. Mid￾dle: The student model is fully fine-tuned to predict consistent novel views G(Iocc), supervised by a denoising loss Ldenoise. Bot￾tom: The predicted six-view images are fe… view at source ↗
Figure 3
Figure 3. Qualitative de-occlusion results on diverse objects. Each triplet shows (left) the original image, (middle) the occluded input, [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of de-occlusion results on the Occ-LVIS benchmark. We compare 3DRecon [ [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative 3D reconstruction results using predicted [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Sign in 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. RelaxFlow: Text-Driven Amodal 3D Generation

    cs.CV 2026-03 conditional novelty 6.5 of 10

    A training-free dual-branch flow method uses multi-prior consensus and attention-logit low-pass relaxation to text-steer occluded 3D geometry while preserving the observed image.

  2. Axolotl3D: a Unified Framework for Faithful 3D Shape Completion

    cs.CV 2026-07 conditional novelty 5.0 of 10

    A single diffusion model, fine-tuned from Hunyuan3D, completes 3D shapes from any mix of images, masks, cameras, and point clouds, reaching state-of-the-art or competitive accuracy on Toys4K and OmniObject3D.

  3. XSpecMesh: Quality-Preserving Auto-Regressive Mesh Generation Acceleration via Multi-Head Speculative Decoding

    cs.GR 2025-07 conditional novelty 5.0 of 10

    XSpecMesh speeds up auto-regressive mesh generation by about 1.7x using multi-head speculative decoding with cross-attention heads and a probability threshold verification, while keeping output quality close to the ba...

Reference graph

Works this paper leans on

51 extracted references · 31 canonical work pages · cited by 3 Pith papers

  1. [1]

    Zero-1-to-3: Zero-shot one image to 3d object,

    R. Liu, R. Wu, B. Van Hoorick, P. Tokmakov, S. Zakharov, and C. V ondrick, “Zero-1-to-3: Zero-shot one image to 3d object,” in Proceedings of the IEEE/CVF international con- ference on computer vision, pp. 9298–9309, 2023

    work page 2023

  2. [2]

    Zero123++: a single image to con- sistent multi-view diffusion base model,

    R. Shi, H. Chen, Z. Zhang, M. Liu, C. Xu, X. Wei, L. Chen, C. Zeng, and H. Su, “Zero123++: a single image to con- sistent multi-view diffusion base model,” arXiv preprint arXiv:2310.15110, 2023

    arXiv 2023

  3. [3]

    Instantmesh: Efficient 3d mesh generation from a single image with sparse-view large reconstruction models,

    J. Xu, W. Cheng, Y . Gao, X. Wang, S. Gao, and Y . Shan, “Instantmesh: Efficient 3d mesh generation from a single image with sparse-view large reconstruction models,” arXiv preprint arXiv:2404.07191, 2024

    arXiv 2024

  4. [4]

    Mvdream: Multi-view diffusion for 3d generation,

    Y . Shi, P. Wang, J. Ye, M. Long, K. Li, and X. Yang, “Mvdream: Multi-view diffusion for 3d generation,” arXiv preprint arXiv:2308.16512, 2023

    arXiv 2023

  5. [5]

    pix2gestalt: Amodal segmenta- tion by synthesizing wholes,

    E. Ozguroglu, R. Liu, D. Sur ´ıs, D. Chen, A. Dave, P. Tok- makov, and C. V ondrick, “pix2gestalt: Amodal segmenta- tion by synthesizing wholes,” in 2024 IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR) , pp. 3931–3940, IEEE Computer Society, 2024

    work page 2024

  6. [6]

    Amodal completion via pro- gressive mixed context diffusion,

    K. Xu, L. Zhang, and J. Shi, “Amodal completion via pro- gressive mixed context diffusion,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 9099–9109, 2024

    work page 2024

  7. [7]

    Generalizable 3d scene reconstruction via divide and conquer from a single view,

    A. Dogaru, M. ¨Ozer, and B. Egger, “Generalizable 3d scene reconstruction via divide and conquer from a single view,” arXiv preprint arXiv:2404.03421, 2024

    arXiv 2024

  8. [8]

    Self-supervised scene de-occlusion,

    X. Zhan, X. Pan, B. Dai, Z. Liu, D. Lin, and C. C. Loy, “Self-supervised scene de-occlusion,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 3784–3792, 2020

    work page 2020

Show all 51 references
  1. [9]

    Amodal ground truth and completion in the wild,

    G. Zhan, C. Zheng, W. Xie, and A. Zisserman, “Amodal ground truth and completion in the wild,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 28003–28013, 2024

    2024

  2. [10]

    Synergyamodal: Deocclude anything with text control,

    X. Li, C. Yi, J. Lai, M. Lin, Y . Qu, S. Zhang, and L. Cao, “Synergyamodal: Deocclude anything with text control,” arXiv preprint arXiv:2504.19506, 2025

    2025 arXiv

  3. [11]

    3d- r2n2: A unified approach for single and multi-view 3d object reconstruction,

    C. B. Choy, D. Xu, J. Gwak, K. Chen, and S. Savarese, “3d- r2n2: A unified approach for single and multi-view 3d object reconstruction,” in Computer vision–ECCV 2016: 14th Eu- ropean conference, amsterdam, the netherlands, October 11- 14, 2016, proceedings, part VIII 14 , pp. ...

    2016

  4. [12]

    Kinectfusion: Real-time dense surface map- ping and tracking,

    R. A. Newcombe, S. Izadi, O. Hilliges, D. Molyneaux, D. Kim, A. J. Davison, P. Kohi, J. Shotton, S. Hodges, and A. Fitzgibbon, “Kinectfusion: Real-time dense surface map- ping and tracking,” in 2011 10th IEEE international sym- posium on mixed and augmented reality, pp. 127–13...

    2011

  5. [13]

    Deepsdf: Learning continuous signed dis- tance functions for shape representation,

    J. J. Park, P. Florence, J. Straub, R. Newcombe, and S. Lovegrove, “Deepsdf: Learning continuous signed dis- tance functions for shape representation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 165–174, 2019

    2019

  6. [14]

    Nerf: Representing scenes as neural radiance fields for view synthesis,

    B. Mildenhall, P. P. Srinivasan, M. Tancik, J. T. Barron, R. Ramamoorthi, and R. Ng, “Nerf: Representing scenes as neural radiance fields for view synthesis,” Communications of the ACM, vol. 65, no. 1, pp. 99–106, 2021

    2021

  7. [15]

    Rip-nerf: Learn- ing rotation-invariant point-based neural radiance field for fine-grained editing and compositing,

    Y . Wang, J. Wang, Y . Qu, and Y . Qi, “Rip-nerf: Learn- ing rotation-invariant point-based neural radiance field for fine-grained editing and compositing,” in Proceedings of the 2023 ACM international conference on multimedia retrieval, pp. 125–134, 2023

    2023

  8. [16]

    Sg-nerf: Semantic-guided point- based neural radiance fields,

    Y . Qu, Y . Wang, and Y . Qi, “Sg-nerf: Semantic-guided point- based neural radiance fields,” in 2023 IEEE International Conference on Multimedia and Expo (ICME) , pp. 570–575, IEEE, 2023

    2023

  9. [17]

    Nerf-dets: Enhancing multi-view 3d object detection with sampling- adaptive network of continuous nerf-based representation,

    C. Huang, X. Li, S. Zhang, L. Cao, and R. Ji, “Nerf-dets: Enhancing multi-view 3d object detection with sampling- adaptive network of continuous nerf-based representation,” arXiv e-prints, pp. arXiv–2404, 2024

    2024

  10. [18]

    pixelnerf: Neu- ral radiance fields from one or few images,

    A. Yu, V . Ye, M. Tancik, and A. Kanazawa, “pixelnerf: Neu- ral radiance fields from one or few images,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 4578–4587, 2021

    2021

  11. [19]

    3d gaussian splatting for real-time radiance field rendering.,

    B. Kerbl, G. Kopanas, T. Leimk ¨uhler, and G. Drettakis, “3d gaussian splatting for real-time radiance field rendering.,” ACM Trans. Graph., vol. 42, no. 4, pp. 139–1, 2023

    2023

  12. [20]

    Goi: Find 3d gaussians of interest with an optimizable open-vocabulary semantic-space hyperplane,

    Y . Qu, S. Dai, X. Li, J. Lin, L. Cao, S. Zhang, and R. Ji, “Goi: Find 3d gaussians of interest with an optimizable open-vocabulary semantic-space hyperplane,” in Proceed- ings of the 32nd ACM International Conference on Multi- media, pp. 5328–5337, 2024

    2024

  13. [21]

    Evolving high-quality rendering and reconstruction in a unified framework with contribution-adaptive regular- ization,

    Y . Shen, Z. Zhang, X. Li, Y . Qu, Y . Lin, S. Zhang, and L. Cao, “Evolving high-quality rendering and reconstruction in a unified framework with contribution-adaptive regular- ization,” arXiv preprint arXiv:2503.00881, 2025

    2025 arXiv

  14. [22]

    Look at the sky: Sky-aware efficient 3d gaussian splatting in the wild,

    Y . Wang, J. Wang, R. Gao, Y . Qu, W. Duan, S. Yang, and Y . Qi, “Look at the sky: Sky-aware efficient 3d gaussian splatting in the wild,” IEEE Transactions on Visualization and Computer Graphics, 2025

    2025

  15. [23]

    Wildseg3d: Segment any 3d objects in the wild from 2d images,

    Y . Guo, J. Hu, Y . Qu, and L. Cao, “Wildseg3d: Segment any 3d objects in the wild from 2d images,” arXiv preprint arXiv:2503.08407, 2025

    2025 arXiv

  16. [24]

    Training- free hierarchical scene understanding for gaussian splatting with superpoint graphs,

    S. Dai, Y . Qu, Z. Li, X. Li, S. Zhang, and L. Cao, “Training- free hierarchical scene understanding for gaussian splatting with superpoint graphs,” arXiv preprint arXiv:2504.13153 , 2025

    2025 arXiv

  17. [25]

    Dream- fusion: Text-to-3d using 2d diffusion,

    B. Poole, A. Jain, J. T. Barron, and B. Mildenhall, “Dream- fusion: Text-to-3d using 2d diffusion,” arXiv preprint arXiv:2209.14988, 2022

    2022 arXiv

  18. [26]

    Director3d: Real-world camera trajectory and 3d scene generation from text,

    X. Li, Z. Lai, L. Xu, Y . Qu, L. Cao, S. Zhang, B. Dai, and R. Ji, “Director3d: Real-world camera trajectory and 3d scene generation from text,”Advances in Neural Information Processing Systems, vol. 37, pp. 75125–75151, 2024

    2024

  19. [27]

    Drag your gaussian: Effective drag-based editing with score distillation for 3d gaussian splatting,

    Y . Qu, D. Chen, X. Li, X. Li, S. Zhang, L. Cao, and R. Ji, “Drag your gaussian: Effective drag-based editing with score distillation for 3d gaussian splatting,” arXiv preprint arXiv:2501.18672, 2025

    2025 arXiv

  20. [28]

    Prolificdreamer: High-fidelity and diverse text-to-3d gener- ation with variational score distillation,

    Z. Wang, C. Lu, Y . Wang, F. Bao, C. Li, H. Su, and J. Zhu, “Prolificdreamer: High-fidelity and diverse text-to-3d gener- ation with variational score distillation,” Advances in Neu- ral Information Processing Systems, vol. 36, pp. 8406–8441, 2023

    2023

  21. [29]

    Euler spiral for shape completion,

    B. B. Kimia, I. Frankel, and A.-M. Popescu, “Euler spiral for shape completion,”International journal of computer vision, vol. 54, no. 1, pp. 159–182, 2003

    2003

  22. [30]

    A contour completion model for augmenting surface reconstructions,

    N. Silberman, L. Shapira, R. Gal, and P. Kohli, “A contour completion model for augmenting surface reconstructions,” in Computer Vision–ECCV 2014: 13th European Confer- ence, Zurich, Switzerland, September 6-12, 2014, Proceed- ings, Part III 13, pp. 488–503, Springer, 2014

    2014

  23. [31]

    Face de-occlusion with deep cascade guidance learning,

    N. Zhang, N. Liu, J. Han, K. Wan, and L. Shao, “Face de-occlusion with deep cascade guidance learning,” IEEE Transactions on Multimedia, vol. 25, pp. 3217–3229, 2022

    2022

  24. [32]

    Visualiz- ing the invisible: Occluded vehicle segmentation and recov- ery,

    X. Yan, F. Wang, W. Liu, Y . Yu, S. He, and J. Pan, “Visualiz- ing the invisible: Occluded vehicle segmentation and recov- ery,” in Proceedings of the IEEE/CVF International Confer- ence on Computer Vision, pp. 7618–7627, 2019

    2019

  25. [33]

    Hu- man de-occlusion: Invisible perception and recovery for hu- mans,

    Q. Zhou, S. Wang, Y . Wang, Z. Huang, and X. Wang, “Hu- man de-occlusion: Invisible perception and recovery for hu- mans,” inProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pp. 3691–3701, 2021

    2021

  26. [34]

    How to make a pizza: Learning a com- positional layer-based gan model,

    D. P. Papadopoulos, Y . Tamaazousti, F. Ofli, I. Weber, and A. Torralba, “How to make a pizza: Learning a com- positional layer-based gan model,” in proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8002–8011, 2019

    2019

  27. [35]

    A diffusion-based framework for oc- cluded object movement,

    Z.-P. Duan, J. Zhang, S. Liu, Z. Lin, C.-L. Guo, D. Zou, J. Ren, and C. Li, “A diffusion-based framework for oc- cluded object movement,” in Proceedings of the AAAI Con- ference on Artificial Intelligence , vol. 39, pp. 2816–2824, 2025

    2025

  28. [36]

    Tuning-free amodal seg- mentation via the occlusion-free bias of inpainting models,

    J. J. Lee, B. Benes, and R. A. Yeh, “Tuning-free amodal seg- mentation via the occlusion-free bias of inpainting models,” arXiv preprint arXiv:2503.18947, 2025

    2025 arXiv

  29. [37]

    Deep depth completion of a single rgb-d image,

    Y . Zhang and T. Funkhouser, “Deep depth completion of a single rgb-d image,” in Proceedings of the IEEE Confer- ence on Computer Vision and Pattern Recognition (CVPR) , pp. 175–185, 2018

    2018

  30. [38]

    Sparse-to-dense: Depth prediction from sparse depth samples and a single im- age,

    F. Ma, G. Cavalheiro, and S. Karaman, “Sparse-to-dense: Depth prediction from sparse depth samples and a single im- age,” in 2018 IEEE International Conference on Robotics and Automation (ICRA), pp. 1–8, IEEE, 2018

    2018

  31. [39]

    Video based reconstruction of 3d people models,

    T. Alldieck, M. A. Magnor, W. Xu, C. Theobalt, and G. Pons- Moll, “Video based reconstruction of 3d people models,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 8387–8397, 2018

    2018

  32. [40]

    Expressive body capture: 3d hands, face, and body from a single image,

    G. Pavlakos, V . Choutas, N. Ghorbani, T. Bolkart, A. A. Os- man, D. Tzionas, and M. J. Black, “Expressive body capture: 3d hands, face, and body from a single image,” in Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 10975–10985, 2019

    2019

  33. [41]

    Chrome: Clothed human reconstruction with occlusion-resilience and multiview-consistency from a single image,

    A. Dutta, M. Zheng, Z. Gao, B. Planche, A. Choudhuri, T. Chen, A. K. Roy-Chowdhury, and Z. Wu, “Chrome: Clothed human reconstruction with occlusion-resilience and multiview-consistency from a single image,” arXiv preprint arXiv:2503.15671, 2025

    2025

  34. [42]

    Occfu- sion: Rendering occluded humans with generative diffusion priors,

    A. Sun, T. Xiang, S. Delp, L. Fei-Fei, and E. Adeli, “Occfu- sion: Rendering occluded humans with generative diffusion priors,” arXiv preprint arXiv:2407.00316, 2024

    2024 arXiv

  35. [43]

    Slice3d: Multi-slice occlusion-revealing single view 3d reconstruction,

    Y . Wang, W. Lira, W. Wang, A. Mahdavi-Amiri, and H. Zhang, “Slice3d: Multi-slice occlusion-revealing single view 3d reconstruction,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 9881–9891, 2024

    2024

  36. [44]

    Sail-vos: Semantic amodal instance level video object segmentation-a synthetic dataset and baselines,

    Y .-T. Hu, H.-S. Chen, K. Hui, J.-B. Huang, and A. G. Schwing, “Sail-vos: Semantic amodal instance level video object segmentation-a synthetic dataset and baselines,” in Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pp. 3105–3115, 2019

    2019

  37. [45]

    Segment anything,

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y . Lo, et al., “Segment anything,” in Proceedings of the IEEE/CVF international conference on computer vision , pp. 4015– 4026, 2023

    2023

  38. [46]

    Structured 3d latents for scalable and versatile 3d generation,

    J. Xiang, Z. Lv, S. Xu, Y . Deng, R. Wang, B. Zhang, D. Chen, X. Tong, and J. Yang, “Structured 3d latents for scalable and versatile 3d generation,” in Proceedings of the Computer Vi- sion and Pattern Recognition Conference, pp. 21469–21480, 2025

    2025

  39. [47]

    Hunyuan3d 2.0: Scal- ing diffusion models for high resolution textured 3d assets generation,

    Z. Zhao, Z. Lai, Q. Lin, Y . Zhao, H. Liu, S. Yang, Y . Feng, M. Yang, S. Zhang, X. Yang, et al., “Hunyuan3d 2.0: Scal- ing diffusion models for high resolution textured 3d assets generation,” arXiv preprint arXiv:2501.12202, 2025

    2025 arXiv

  40. [48]

    Objaverse: A universe of annotated 3d objects,

    M. Deitke, D. Schwenk, J. Salvador, L. Weihs, O. Michel, E. VanderBilt, L. Schmidt, K. Ehsani, A. Kembhavi, and A. Farhadi, “Objaverse: A universe of annotated 3d objects,” in Proceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pp. 13142–13153, 2023

    2023

  41. [49]

    Gans trained by a two time-scale update rule converge to a local nash equilibrium,

    M. Heusel, H. Ramsauer, T. Unterthiner, B. Nessler, and S. Hochreiter, “Gans trained by a two time-scale update rule converge to a local nash equilibrium,” Advances in neural information processing systems, vol. 30, 2017

    2017

  42. [50]

    Demystifying mmd gans,

    M. Bi ´nkowski, D. J. Sutherland, M. Arbel, and A. Gretton, “Demystifying mmd gans,” arXiv preprint arXiv:1801.01401, 2018

    2018 arXiv

  43. [51]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark,et al., “Learning transferable visual models from natural language supervision,” in International conference on machine learn- ing, pp. 8748–8763, PmLR, 2021

    2021

Pith tools

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