Pith. sign in

REVIEW 3 major objections 8 minor 53 references

Amodal Depth Anything: Amodal Depth Estimation in the Wild

T0 review · 3 major / 8 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that amodal depth — the depth of object parts hidden behind occluders — can be learned from real images by reposing the task as relative depth and training on a large composited dataset with Depth Anything V2 as teacher.

desk verdict Useful dataset and two working models for amodal relative depth, but the headline numbers measure agreement with the teacher, not physical depth. read the letter →

arxiv 2412.02336 v1 pith:66SWSWVE submitted 2024-12-03 cs.CV

Zhenyu Li , Mykola Lavreniuk , Jian Shi , Shariq Farooq Bhat , Peter Wonka This is my paper · ORCID
classification cs.CV
keywords amodaldepthestimationrelativepredictionADIWdatasetoccludedgeometryscale-and-shiftalignmentflowmatchingAnythingV2inpainting
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 establish that the depth of occluded, invisible parts of objects can be predicted from a single natural image, and that this amodal depth task is best posed as relative depth rather than metric depth. To support that claim it introduces ADIW, a 564K-image dataset built by compositing segmented objects onto natural backgrounds and labeling them with depth from a large pre-trained relative-depth model, then aligning scales so the hidden geometry is consistent with visible geometry. On this dataset, the deterministic Amodal-DA V2 model and the generative Amodal-DepthFM model both beat prior amodal depth methods and inpainting-based alternatives, with the best model reporting a 27.4% RMSE improvement over the previous state of the art. A sympathetic reader would care because, if true, it means geometry behind occluders is learnable from real-world image statistics rather than requiring hand-built synthetic scenes.

What carries the argument

The scale-and-shift alignment is the load-bearing mechanism: for each composited training pair, Eqs. 1-2 fit the background image's depth to the observed image's visible depth with least-squares scale $s$ and shift $t$, producing $D_{\mathrm{aligned}} = sD_b + t$ as ground truth. This transfers the pre-trained model's depth for the hidden part into the composite. The second mechanism is the guidance-channel modification: a zero-initialized parallel convolution adds the observation depth map $D_o$ and amodal mask $M_a$ to the encoder, so fine-tuning starts from pre-trained weights and the model learns to condition on those signals. Together these convert a monocular depth foundation model into an amodal depth predictor with minimal architecture change.

What would settle it

Take real scenes with a known object partially hidden behind another object, measure the true depth of the hidden surface after removing the occluder, for example with a depth sensor or multi-view reconstruction, and compare those measurements with both the model predictions on the occluded image and ADIW's assembled labels. If the physical hidden depths disagree systematically with the model or with the aligned Depth Anything V2 labels, the reported accuracy would be measuring agreement with a teacher prior rather than true amodal geometry.

Watch

Extended reading notes

Core claim

The central claim is that relative amodal depth can be learned in the wild: given an image, its observed depth map, and an amodal mask marking where an object extends behind an occluder, a model fine-tuned from Depth Anything V2 can output a depth map that includes the occluded region, and this transfers to real images better than previous metric-depth amodal models. The dataset construction is the enabling move: rather than measuring hidden depth directly, which no sensor can do at scale, the pipeline uses the depth a strong relative-depth model assigns to an un-occluded image of the same object, aligned to the visible depth by a global scale and shift. The paper further claims that both the deterministic and generative adaptations work, that object-level supervision helps, and that the guidance channels carrying the observed depth and the amodal mask account for most of the performance gain.

Load-bearing premise

The load-bearing premise is that the depth a large pre-trained model assigns to an object in an un-occluded background image is the true depth of that object's hidden geometry in the composite, after a global scale-and-shift fit on visible pixels.

Editorial extensions

If this is right

  • If the central claim holds, relative-depth amodal estimation can be trained from real images at scale without any physical sensor for occluded depth.
  • The reported results imply that the ADIW benchmark separates methods on occluded-region accuracy, with Amodal-DA V2-L reducing RMSE by 27.4% over the previous state of the art.
  • Ablations imply that both the observed-depth and amodal-mask guidance channels are necessary, since removing either degrades accuracy.
  • The generative variant implies that multiple plausible hidden geometries can be sampled for one scene, not just a single deterministic prediction.
  • The comparison with inpainting-based baselines implies that direct depth regression is more reliable than inpainting-then-depth-estimation pipelines for occluded geometry.

Reading between the lines

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

  • Beyond the paper, the same compositing-and-alignment pipeline could be reused to train amodal models for surface normals, segmentation, or RGB inpainting by swapping the teacher model.
  • Because ADIW labels are inherited from Depth Anything V2, the benchmark's improvements are partly a measure of consistency with that model's geometric prior; a physical-ground-truth test could change the ranking.
  • The framework's real-world value depends on upstream amodal segmentation quality, since the limitation section notes that inaccurate or ambiguous masks propagate errors into depth predictions.
  • A direct evaluation with depth sensors on partially occluded real objects would separate learned geometric priors from alignment artifacts; the paper does not provide such an evaluation.
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 / 8 minor

Summary. The paper proposes relative amodal depth estimation "in the wild": given an RGB image, an observation depth map Do, and an amodal mask Ma, predict depth for the occluded part of the target object. The authors introduce ADIW, a 564K-pair dataset built by compositing an occluder over a background image taken from SAM/SA-1B segmentations; DA V2 (ViT-G) produces depth for both the composite and the background, and a per-image scale-and-shift fit on visible pixels (Eqs. 1-2) generates the occluded-region ground truth. Two models are fine-tuned from pre-trained depth models with minimal architectural changes: Amodal-DA V2, a DA V2 ViT with an extra zero-initialized guidance convolution and layer norm in the DPT head, trained with the scale-invariant log loss; and Amodal-DepthFM, a conditional flow-matching model extended with Do and Ma guidance channels, with scale-and-shift alignment applied at inference. On the ADIW validation split, the best model (Amodal-DA V2-L) reports RMSE 3.418 versus 4.712 for the retrained previous-SoTA baseline Jo et al. (a 27.4% relative improvement), and the abstract claims a 69.5% accuracy improvement over the previous SoTA. Ablations show that the guidance channels, whole-object supervision, and (for DepthFM) inference-time alignment each help. The paper also shows qualitative zero-shot results and 3D mesh reconstructions of occluded objects.

Significance. The main value of the paper is in dataset construction and model engineering. ADIW appears to be the first large-scale training resource for relative amodal depth, the compositing pipeline is genuinely scalable, the two model adaptations are simple and clearly described, and the ablations in Tables 2-3 are informative. The paper is also transparent that no existing sensor can directly measure occluded-region depth (Secs. 1 and 3.2). However, the significance is conditional: the ADIW validation labels are generated by DA V2 itself (Sec. 3.2, Eqs. 1-2), the best model is initialized from DA V2 (Sec. 4.2), and the reported state of the art therefore currently demonstrates accurate reproduction of DA V2's hidden-geometry predictions rather than physical amodal depth. The contribution would become a clear accept with external validation against true occluded geometry (for example, synthetic benchmarks with rendered ground truth or a real-sensor split), or with the claims explicitly reframed as teacher-agreement results. As written, the headline numbers in the abstract and Sec. 4.3 overstate what has been established.

major comments (3)
  1. [Secs. 1, 3.2, 4.3; Eqs. (1)-(2)] The ADIW ground truth for occluded regions is defined as DA V2's relative depth on the un-occluded background image, aligned to the observation depth with a per-image scale-and-shift fit (Eqs. 1-2), and the 4K-sample validation split is produced by the same pipeline. The best model, Amodal-DA V2-L, is initialized from DA V2 and fine-tuned on these labels (Sec. 4.2). The headline numbers in Table 1 (the 27.4% RMSE improvement over Jo et al. retrained on ADIW, and the abstract's 69.5% accuracy claim) therefore measure how faithfully the fine-tuned model reproduces DA V2's own hidden-geometry estimates, and a DA V2-initialized model has an inherent advantage over baselines that must learn the teacher's prior from the labels alone. The paper's own statements that no device can capture occluded-region ground truth (Secs. 1 and 3.2) make this a load-bearing limitation rather than a side remark. This does not invalidate ADIW as a training resource, but the claim of amodal depth estimation "in the wild" is not yet established. Please reframe the evaluation and SoTA claims as measuring agreement with the teacher prior, and add external validation of physical accuracy, for example zero-shot evaluation on Amodal-SynthDrive and Amodal-3D-FRONT (which have rendered ground-truth depth behind occluders) or on a split where a real RGB-D sensor measures the background depth before a physical occluder is placed.
  2. [Sec. 4.1 and Table 1] The evaluation protocol does not specify how predicted relative depth maps are brought into the scale of the ground truth before RMSE, log10, and delta are computed. The ground truth is defined in the canonical scale of Do (normalized to [0,1] and then scale-shifted per image, Eqs. 1-2), while the baselines marked with a double dagger are explicitly scale-shift aligned to the observation depth at inference and Amodal-DA V2 is not. Since all three metrics are scale-sensitive, the absolute numbers in Table 1 cannot be independently reproduced without stating the normalization applied to each method's raw output (for example, whether predictions are re-normalized to [0,1], fitted to Do, or used as-is). Please make the protocol explicit and show that the reported differences are not an artifact of scale matching; this is needed for the 27.4% SoTA claim to be verifiable.
  3. [Abstract and Sec. 4.3] The abstract's 69.5% accuracy claim is not consistent with the body's 27.4% RMSE claim. The 27.4% figure is reproducible from Table 1 against the aligned retrained previous-SoTA row (Jo et al. with double-dagger alignment, RMSE 4.712 versus 3.418). The 69.5% figure is not: relative to the aligned Jo et al. row, the delta accuracy improves by about 5.6% (89.177 to 94.179), whereas the 69.5% value matches the relative delta improvement against the unaligned Jo et al. row (55.545 to 94.179), which is the weaker configuration the table itself marks as not the recommended comparison. Please state explicitly which baseline and which metric the abstract's headline number refers to, and recompute it against the same configuration used for the SoTA comparison in Sec. 4.3.
minor comments (8)
  1. [Sec. 3.3.1] The scale-invariant loss Lsi is written with constants alpha and lambda, but alpha is never defined; with lambda = 0.85 the loss is only partially scale-invariant, which is directly relevant to the scale-normalization question raised above, so please define all constants and state the resulting scale behavior of the training target.
  2. [Sec. 3.3.2] The flow-matching objective is written as min over theta of the expectation of the norm of v_theta(t, phi_t(x0)) minus (x1 - x0), but the norm subscript is missing, the marginalization over p(x0) is not defined, and the notation mixes the conditioning code z with phi_t(x|z); please clean up the notation.
  3. [Sec. 4.3 and Table 1] The phrase "previous SoTA amodal depth model" is applied to Jo et al. [16] even though ADIW is a new benchmark and the comparison is run on a model retrained on ADIW; please reword to make clear that the baseline is the best existing amodal depth architecture retrained on the new dataset.
  4. [Abstract and throughout] The model name alternates between "Amodal-DAV2" and "Amodal-DA V2"; please use one consistent form throughout the paper.
  5. [References] References [48] and [49] are the same paper (Zhan et al., "Self-supervised scene de-occlusion," CVPR 2020) and are cited as two distinct entries; please merge them.
  6. [Sec. 4.1] The metrics "log 10 error" and "accuracy under the threshold delta" are never given formulas; since these are scale-sensitive for relative depth, please define them (for example, mean absolute log10 error, and the fraction of pixels satisfying max(y/y_hat, y_hat/y) < 1.25).
  7. [Appendix C] The limitation section discusses dependence on the amodal mask and a decline in detail-capturing ability, but it does not mention that the benchmark ground truth is teacher-generated; adding this caveat would help future users interpret ADIW scores correctly.
  8. [Sec. 3.2] The dataset is described as "real-world data" in one place and "realistic synthetically generated datasets" in another; since the compositing step is synthetic, the wording should be consistent to avoid confusion about what is real versus what is composed.

Circularity Check

1 steps flagged · score 6.0 of 10

ADIW ground truth is Depth Anything V2's own output, and Amodal-DA V2 is initialized from that same model, so the reported SoTA measures self-consistency with the teacher rather than independent amodal depth accuracy.

  1. self definitional [Sec. 3.2, Eqs. (1)-(2); Sec. 3.3.1; Sec. 4.3]
    "Both Io and Ib are then processed through the Depth Anything V2 model [46] (ViT-G) to obtain relative depth maps Do and Db ... The aligned depth map Daligned is then calculated as: Daligned = sDb + t, (2), serving as the ground-truth map for model training."

    The supervision signal for occluded regions is literally DA V2's depth estimate on the background image, rescaled by s,t fit to visible pixels. Amodal-DA V2 is initialized from DA V2 and fine-tuned on these same labels, and the ADIW validation split is produced by the same pipeline. Consequently, the 27.4% RMSE improvement (and the 69.5% accuracy claim) largely measures how faithfully the fine-tuned model reproduces DA V2's prior for hidden geometry, not agreement with any independent ground truth. The paper itself states that no device can capture occluded-part depth, so the benchmark is self-consistent by construction.

full rationale

The paper's core contribution is self-contained as a training pipeline: a 564K-pair dataset with composited objects, two fine-tuned architectures, and ablations. There is no load-bearing self-citation chain: DA V2 and DepthFM are external pre-trained models, and the authors' own prior depth papers are not used to justify the central amodal claim. The circularity lies in the evaluation: Eqs. (1)-(2) define the occluded-region ground truth as an affine-aligned version of DA V2's background depth map, the same model family from which Amodal-DA V2 is initialized and fine-tuned. The validation set is generated by the identical pipeline, so the headline improvements over baselines measure fit to DA V2's own depth priors. This does not make the dataset useless; it is a valid training resource for learning a DA V2-consistent amodal prior. However, it does mean the paper has not established accuracy against physically verified amodal depth, and the central SoTA claim reduces to self-consistency with the label generator, yielding a score of 6.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim rests on pseudo-labels produced by Depth Anything V2; no true occluded-depth sensor exists. The only invented artifact is the ADIW dataset, which is a dataset, not a postulated entity. Per-image scale and shift are the only fitted numbers that materially affect all labels and several reported scores.

free parameters (2)
  • scale s (per-image alignment) = fitted per image pair
    Least-squares solution of Eq. (1); used to build all ADIW labels and to align predictions at inference.
  • shift t (per-image alignment) = fitted per image pair
    Companion to s; same rationale.
assumptions (4)
  • domain assumption The depth of an occluded background object equals the depth predicted by Depth Anything V2 for the complete object, up to a global scale and shift.
    Invoked in Sec. 3.2 (Eqs. 1-2). No physical measurement supports this; the same model family (DA V2) also initializes the best-performing framework.
  • domain assumption The pix2gestalt heuristic ([27]) correctly identifies complete objects and their amodal extents in SA-1B.
    Used in Sec. 3.2 to form the whole-object dataset; incorrect amodal extents would corrupt labels.
  • domain assumption A single global scale-and-shift alignment is sufficient to reconcile relative depth maps in a composite image.
    Eq. (2) assumes non-local distortion errors are negligible; local errors in DA V2 depth are not corrected.
  • domain assumption The ADIW validation split is a valid proxy for amodal depth accuracy in the wild.
    Sec. 4.1/4.3; no external ground truth or cross-benchmark evaluation is provided, and zero-shot claims are qualitative only.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Amodal Depth Anything: Amodal Depth Estimation in the Wild." pith.science (2026). https://pith.science/paper/66SWSWVE

@misc{pith2026241202336,
  author       = {Pith},
  title        = {Pith review of: Amodal Depth Anything: Amodal Depth Estimation in the Wild},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/66SWSWVE}},
  note         = {Machine review of arXiv:2412.02336}
}
read the original abstract

Amodal depth estimation aims to predict the depth of occluded (invisible) parts of objects in a scene. This task addresses the question of whether models can effectively perceive the geometry of occluded regions based on visible cues. Prior methods primarily rely on synthetic datasets and focus on metric depth estimation, limiting their generalization to real-world settings due to domain shifts and scalability challenges. In this paper, we propose a novel formulation of amodal depth estimation in the wild, focusing on relative depth prediction to improve model generalization across diverse natural images. We introduce a new large-scale dataset, Amodal Depth In the Wild (ADIW), created using a scalable pipeline that leverages segmentation datasets and compositing techniques. Depth maps are generated using large pre-trained depth models, and a scale-and-shift alignment strategy is employed to refine and blend depth predictions, ensuring consistency in ground-truth annotations. To tackle the amodal depth task, we present two complementary frameworks: Amodal-DAV2, a deterministic model based on Depth Anything V2, and Amodal-DepthFM, a generative model that integrates conditional flow matching principles. Our proposed frameworks effectively leverage the capabilities of large pre-trained models with minimal modifications to achieve high-quality amodal depth predictions. Experiments validate our design choices, demonstrating the flexibility of our models in generating diverse, plausible depth structures for occluded regions. Our method achieves a 69.5% improvement in accuracy over the previous SoTA on the ADIW dataset.

Figures

Figures reproduced from arXiv: 2412.02336 by the authors.

Figure 1
Figure 1. Amodal Depth Estimation in the Wild. For each image, we present the general depth estimation result alongside our amodal depth estimation with the target object outlined in black. Our model demonstrates strong generalization across diverse scenes for accurate depth estimation for occluded parts of objects. Best viewed in color. Abstract Amodal depth estimation aims to predict the depth of occluded (invisible) parts … view at source ↗
Figure 2
Figure 2. Amodal Depth Estimation Pipeline. Given an input image, users can generate the amodal mask for the depth estimator in two ways: (1) Model Heuristics: click the target object, apply SAM [20] to generate modal mask, then use amodal segmentation methods to estimate amodal mask, (2) Human Heuristics: man￾ually draw the amodal mask. Our model estimates amodal depth based on original observation image Io, the observed dep… view at source ↗
Figure 4
Figure 4. Amodal-DAV2 Framework Structure. Amodal-DAV2 modifies the DAV2 image encoder to take additional guidance channels along with RGB. part of the background object in Do and Db, respectively, and N denotes the total number of valid pixels in the visible mask Mvis of the background object. The aligned depth map Daligned is then calculated as: Daligned = sDb + t, (2) serving as the ground-truth map for model training. Thi… view at source ↗
Figures from the paper (8 more)
Figure 5
Figure 5. Figure 5: Amodal-DepthFM Framework Structure. Amodal￾DepthFM modifies the DepthFM denoising UNet encoder to take additional guidance channels along with RGB latent code. knowledge while enhancing its capacity in amodal depth estimation. The original DAV2 model employs a Vision T…
Figure 6
Figure 6. Figure 6: Qualitative Comparisons on the Validation Set of ADIW. Since Invisible Stitch (IS) [10] and Jo et al. [16] use metric depth estimation models [3], the depth maps are shifted compared with other methods that use relative depth estimators. We evaluate their released mode…
Figure 7
Figure 7. Figure 7: Qualitative Comparisons on Images in the Wild. Invisible Stitch (IS) [10] uses SD-XL [29] with a ground-truth amodal mask and image caption to inpaint occluded areas, while pix2gestalt [27] completes the occluded areas via amodal inpainting. Both methods suffer from in…
Figure 8
Figure 8. Figure 8: Results with Different Mask Guidance. Guided by various amodal masks, our model successfully predicts the corre￾sponding amodal depth maps for the target objects in the image, showcasing flexibility in estimating occluded regions. Amodal-DepthFM produces depth maps wit…
Figure 9
Figure 9. Figure 9: Amodal-DAV2 vs. Amodal-DepthFM. While Amodal￾DAV2 achieves better overall performance, Amodal-DepthFM provides more detailed depth with sharper boundaries. Its gen￾erative nature allows it to synthesize multiple plausible structures. on a generative flow matching parad…
Figure 10
Figure 10. Figure 10: Invisible Stitch for Amodal Depth. Depth Est. Amodal Mask Seg. Depth Est. pix2gestalt Image Observation Depth Amodal Depth Object Mask Object Depth Align [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Pix2gestalt Stitch for Amodal Depth [PITH_FULL_IMAGE:figures/full_fig_p009_11.png]
Figure 12
Figure 12. Figure 12: Reconstructed 3D Mesh for Occluded Object. Blue arrows indicate the target object and red arrows highlight the re￾constructed meshes for occluded parts of objects, respectively. Left: Input image. Middle: Mesh from general depth. Right: Re￾constructed mesh combining g…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 36 canonical work pages

  1. [1]

    Adabins: Depth estimation using adaptive bins

    Shariq Farooq Bhat, Ibraheem Alhashim, and Peter Wonka. Adabins: Depth estimation using adaptive bins. In CVPR, pages 4009–4018, 2021. 1, 3, 4, 5, 6

    work page 2021

  2. [2]

    Localbins: Improving depth estimation by learning local dis- tributions

    Shariq Farooq Bhat, Ibraheem Alhashim, and Peter Wonka. Localbins: Improving depth estimation by learning local dis- tributions. In European Conference on Computer Vision , pages 480–496. Springer, 2022. 3

    work page 2022

  3. [3]

    Zoedepth: Zero-shot trans- fer by combining relative and metric depth

    Shariq Farooq Bhat, Reiner Birkl, Diana Wofk, Peter Wonka, and Matthias M ¨uller. Zoedepth: Zero-shot trans- fer by combining relative and metric depth. arXiv preprint arXiv:2302.12288, 2023. 1, 3, 6

    arXiv 2023

  4. [4]

    Monocular depth estimation: A survey

    Amlaan Bhoi. Monocular depth estimation: A survey. arXiv preprint arXiv:1901.09402, 2019. 1

    arXiv 1901

  5. [5]

    Midas v3.1 – a model zoo for robust monocular relative depth estimation

    Reiner Birkl, Diana Wofk, and Matthias M ¨uller. Midas v3.1 – a model zoo for robust monocular relative depth estimation. arXiv preprint arXiv:2307.14460, 2023. 2, 3, 4

    arXiv 2023

  6. [6]

    Depth pro: Sharp monocular metric depth in less than a second

    Aleksei Bochkovskii, Ama ¨el Delaunoy, Hugo Germain, Marcel Santos, Yichao Zhou, Stephan R Richter, and Vladlen Koltun. Depth pro: Sharp monocular metric depth in less than a second. arXiv preprint arXiv:2410.02073, 2024. 2, 4

    arXiv 2024

  7. [7]

    An image is worth 16x16 words: Trans- formers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, et al. An image is worth 16x16 words: Trans- formers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020. 5

    arXiv 2010

  8. [8]

    Segan: Segmenting and generating the invisible

    Kiana Ehsani, Roozbeh Mottaghi, and Ali Farhadi. Segan: Segmenting and generating the invisible. In CVPR, pages 6144–6153, 2018. 3

    work page 2018

Show all 53 references
  1. [9]

    Depth map prediction from a single image using a multi-scale deep net- work

    David Eigen, Christian Puhrsch, and Rob Fergus. Depth map prediction from a single image using a multi-scale deep net- work. NeurIPS, 27, 2014. 1, 3, 5

    2014

  2. [10]

    Invisible stitch: Generating smooth 3d scenes with depth inpainting

    Paul Engstler, Andrea Vedaldi, Iro Laina, and Christian Rup- precht. Invisible stitch: Generating smooth 3d scenes with depth inpainting. arXiv preprint arXiv:2404.19758, 2024. 2, 3, 6, 7, 8, 9

    2024 arXiv

  3. [11]

    Deep ordinal regression net- work for monocular depth estimation

    Huan Fu, Mingming Gong, Chaohui Wang, Kayhan Bat- manghelich, and Dacheng Tao. Deep ordinal regression net- work for monocular depth estimation. InCVPR, pages 2002– 2011, 2018. 3

    2002

  4. [12]

    Geobench: Benchmarking and analyzing monocular geom- etry estimation models

    Yongtao Ge, Guangkai Xu, Zhiyue Zhao, Libo Sun, Zheng Huang, Yanlong Sun, Hao Chen, and Chunhua Shen. Geobench: Benchmarking and analyzing monocular geom- etry estimation models. arXiv preprint arXiv:2406.12671 ,

    arXiv

  5. [13]

    Depthfm: Fast monocular depth estimation with flow matching

    Ming Gui, Johannes S Fischer, Ulrich Prestel, Pingchuan Ma, Dmytro Kotovenko, Olga Grebenkova, Stefan Andreas Baumann, Vincent Tao Hu, and Bj ¨orn Ommer. Depthfm: Fast monocular depth estimation with flow matching. arXiv preprint arXiv:2403.13788, 2024. 2, 3, 4, 5, 7

    2024 arXiv

  6. [14]

    Tracking any object amodally

    Cheng-Yen Hsieh, Tarasha Khurana, Achal Dave, and Deva Ramanan. Tracking any object amodally. arXiv preprint arXiv:2312.12433, 2023. 3

    2023 arXiv

  7. [15]

    Slide: Single image 3d photography with soft layering and depth-aware inpainting

    Varun Jampani, Huiwen Chang, Kyle Sargent, Abhishek Kar, Richard Tucker, Michael Krainin, Dominik Kaeser, William T Freeman, David Salesin, Brian Curless, et al. Slide: Single image 3d photography with soft layering and depth-aware inpainting. In ICCV, pages 12518–12527, 2021. 3

    2021

  8. [16]

    Occlusion- aware amodal depth estimation for enhancing 3d reconstruc- tion from a single image

    Seong-Uk Jo, Du Yeol Lee, and Chae Eun Rhee. Occlusion- aware amodal depth estimation for enhancing 3d reconstruc- tion from a single image. IEEE Access, 2024. 2, 3, 4, 6, 7, 8

    2024

  9. [17]

    Amodal completion and size constancy in natural scenes

    Abhishek Kar, Shubham Tulsiani, Joao Carreira, and Jiten- dra Malik. Amodal completion and size constancy in natural scenes. In ICCV, pages 127–135, 2015. 3

    2015

  10. [18]

    Repurpos- ing diffusion-based image generators for monocular depth estimation

    Bingxin Ke, Anton Obukhov, Shengyu Huang, Nando Met- zger, Rodrigo Caye Daudt, and Konrad Schindler. Repurpos- ing diffusion-based image generators for monocular depth estimation. In CVPR, pages 9492–9502, 2024. 3

    2024

  11. [19]

    Deep occlusion- aware instance segmentation with overlapping bilayers

    Lei Ke, Yu-Wing Tai, and Chi-Keung Tang. Deep occlusion- aware instance segmentation with overlapping bilayers. In CVPR, pages 4019–4028, 2021. 3

    2021

  12. [20]

    Segment any- thing

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C Berg, Wan-Yen Lo, et al. Segment any- thing. In ICCV, pages 4015–4026, 2023. 2, 4

    2023

  13. [21]

    Evp: Enhanced visual perception us- ing inverse multi-attentive feature refinement and regularized image-text alignment

    Mykola Lavreniuk, Shariq Farooq Bhat, Matthias Muller, and Peter Wonka. Evp: Enhanced visual perception us- ing inverse multi-attentive feature refinement and regularized image-text alignment. In European Conference on Computer Vision Workshops (ECCVW), 2024. 3

    2024

  14. [22]

    Binsformer: Revisiting adaptive bins for monocular depth estimation

    Zhenyu Li, Xuyang Wang, Xianming Liu, and Junjun Jiang. Binsformer: Revisiting adaptive bins for monocular depth estimation. arXiv preprint arXiv:2204.00987, 2022. 1, 3, 4, 6

    2022 arXiv

  15. [23]

    Patch- fusion: An end-to-end tile-based framework for high- resolution monocular metric depth estimation.arXiv preprint arXiv:2312.02284, 2023

    Zhenyu Li, Shariq Farooq Bhat, and Peter Wonka. Patch- fusion: An end-to-end tile-based framework for high- resolution monocular metric depth estimation.arXiv preprint arXiv:2312.02284, 2023. 3

    2023 arXiv

  16. [24]

    Depthformer: Exploiting long-range correlation and local in- formation for accurate monocular depth estimation.Machine Intelligence Research, pages 1–18, 2023

    Zhenyu Li, Zehui Chen, Xianming Liu, and Junjun Jiang. Depthformer: Exploiting long-range correlation and local in- formation for accurate monocular depth estimation.Machine Intelligence Research, pages 1–18, 2023. 3, 4, 6

    2023

  17. [25]

    Humans as light bulbs: 3d hu- man reconstruction from thermal reflection

    Ruoshi Liu and Carl V ondrick. Humans as light bulbs: 3d hu- man reconstruction from thermal reflection. In CVPR, pages 12531–12542, 2023. 3

    2023

  18. [26]

    Single image depth estimation: An overview

    Alican Mertan, Damien Jade Duff, and Gozde Unal. Single image depth estimation: An overview. Digital Signal Pro- cessing, 123:103441, 2022. 1 10

    2022

  19. [27]

    pix2gestalt: Amodal segmentation by synthesizing wholes

    Ege Ozguroglu, Ruoshi Liu, D ´ıdac Sur´ıs, Dian Chen, Achal Dave, Pavel Tokmakov, and Carl V ondrick. pix2gestalt: Amodal segmentation by synthesizing wholes. In CVPR, pages 3931–3940. IEEE Computer Society, 2024. 2, 3, 4, 6, 7, 8, 9

    2024

  20. [28]

    Unidepth: Universal monocular metric depth estimation

    Luigi Piccinelli, Yung-Hsu Yang, Christos Sakaridis, Mattia Segu, Siyuan Li, Luc Van Gool, and Fisher Yu. Unidepth: Universal monocular metric depth estimation. In CVPR, pages 10106–10116, 2024. 2, 4

    2024

  21. [29]

    Sdxl: Improving latent diffusion mod- els for high-resolution image synthesis

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion mod- els for high-resolution image synthesis. arXiv preprint arXiv:2307.01952, 2023. 3, 7, 8, 9

    2023 arXiv

  22. [30]

    Amodal instance segmentation with kins dataset

    Lu Qi, Li Jiang, Shu Liu, Xiaoyong Shen, and Jiaya Jia. Amodal instance segmentation with kins dataset. In CVPR, pages 3014–3023, 2019. 3

    2019

  23. [31]

    Vip-deeplab: Learning visual perception with depth-aware video panoptic segmentation

    Siyuan Qiao, Yukun Zhu, Hartwig Adam, Alan Yuille, and Liang-Chieh Chen. Vip-deeplab: Learning visual perception with depth-aware video panoptic segmentation. In CVPR, pages 3997–4008, 2021. 3

    2021

  24. [32]

    Vi- sion transformers for dense prediction

    Ren ´e Ranftl, Alexey Bochkovskiy, and Vladlen Koltun. Vi- sion transformers for dense prediction. In ICCV, pages 12179–12188, 2021. 5

    2021

  25. [33]

    Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer

    Ren ´e Ranftl, Katrin Lasinger, David Hafner, Konrad Schindler, and Vladlen Koltun. Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer. IEEE TPAMI, 44(3), 2022. 2, 3, 4, 5

    2022

  26. [34]

    Sam 2: Segment anything in images and videos

    Nikhila Ravi, Valentin Gabeur, Yuan-Ting Hu, Ronghang Hu, Chaitanya Ryali, Tengyu Ma, Haitham Khedr, Roman R¨adle, Chloe Rolland, Laura Gustafson, et al. Sam 2: Segment anything in images and videos. arXiv preprint arXiv:2408.00714, 2024. 4

    2024 arXiv

  27. [35]

    360monodepth: High-resolution 360deg monocular depth estimation

    Manuel Rey-Area, Mingze Yuan, and Christian Richardt. 360monodepth: High-resolution 360deg monocular depth estimation. In CVPR, pages 3762–3772, 2022. 5

    2022

  28. [36]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. In CVPR, pages 10684– 10695, 2022. 3

    2022

  29. [37]

    Amodalsynthdrive: A syn- thetic amodal perception dataset for autonomous driving

    Ahmed Rida Sekkat, Rohit Mohan, Oliver Sawade, Elmar Matthes, and Abhinav Valada. Amodalsynthdrive: A syn- thetic amodal perception dataset for autonomous driving. IEEE Robotics and Automation Letters , 2024. 2, 3, 4, 6, 8

    2024

  30. [38]

    3d photography using context-aware layered depth inpainting

    Meng-Li Shih, Shih-Yang Su, Johannes Kopf, and Jia-Bin Huang. 3d photography using context-aware layered depth inpainting. In CVPR, pages 8028–8038, 2020. 2, 3

    2020

  31. [39]

    Alpha- clip: A clip model focusing on wherever you want

    Zeyi Sun, Ye Fang, Tong Wu, Pan Zhang, Yuhang Zang, Shu Kong, Yuanjun Xiong, Dahua Lin, and Jiaqi Wang. Alpha- clip: A clip model focusing on wherever you want. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13019–13029, 2024. 5

    2024

  32. [40]

    Bi- lateral propagation network for depth completion

    Jie Tang, Fei-Peng Tian, Boshi An, Jian Li, and Ping Tan. Bi- lateral propagation network for depth completion. In CVPR, pages 9763–9772, 2024. 3

    2024

  33. [41]

    Neural discrete representation learning

    Aaron Van Den Oord, Oriol Vinyals, et al. Neural discrete representation learning. NeurIPS, 30, 2017. 3

    2017

  34. [42]

    Lrru: Long-short range recurrent updating networks for depth completion

    Yufei Wang, Bo Li, Ge Zhang, Qi Liu, Tao Gao, and Yuchao Dai. Lrru: Long-short range recurrent updating networks for depth completion. In CVPR, pages 9422–9432, 2023. 3

    2023

  35. [43]

    Improving depth completion via depth feature upsampling

    Yufei Wang, Ge Zhang, Shaoqian Wang, Bo Li, Qi Liu, Le Hui, and Yuchao Dai. Improving depth completion via depth feature upsampling. In CVPR, pages 21104–21113, 2024

    2024

  36. [44]

    Tri- perspective view decomposition for geometry-aware depth completion

    Zhiqiang Yan, Yuankai Lin, Kun Wang, Yupeng Zheng, Yufei Wang, Zhenyu Zhang, Jun Li, and Jian Yang. Tri- perspective view decomposition for geometry-aware depth completion. In CVPR, pages 4874–4884, 2024. 3

    2024

  37. [45]

    Depth anything: Unleash- ing the power of large-scale unlabeled data

    Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Ji- ashi Feng, and Hengshuang Zhao. Depth anything: Unleash- ing the power of large-scale unlabeled data. arXiv preprint arXiv:2401.10891, 2024. 1, 2, 3, 4

    2024 arXiv

  38. [46]

    Depth any- thing v2

    Lihe Yang, Bingyi Kang, Zilong Huang, Zhen Zhao, Xiao- gang Xu, Jiashi Feng, and Hengshuang Zhao. Depth any- thing v2. arXiv preprint arXiv:2406.09414, 2024. 2, 3, 4, 7, 9

    2024 arXiv

  39. [47]

    Amodal ground truth and completion in the wild

    Guanqi Zhan, Chuanxia Zheng, Weidi Xie, and Andrew Zis- serman. Amodal ground truth and completion in the wild. In CVPR, pages 28003–28013, 2024. 2, 3

    2024

  40. [48]

    Self-supervised scene de- occlusion

    Xiaohang Zhan, Xingang Pan, Bo Dai, Ziwei Liu, Dahua Lin, and Chen Change Loy. Self-supervised scene de- occlusion. In CVPR, pages 3784–3792, 2020. 2, 3

    2020

  41. [49]

    Self-supervised scene de- occlusion

    Xiaohang Zhan, Xingang Pan, Bo Dai, Ziwei Liu, Dahua Lin, and Chen Change Loy. Self-supervised scene de- occlusion. In CVPR, pages 3784–3792, 2020. 2

    2020

  42. [50]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In ICCV, pages 3836–3847, 2023. 7

    2023

  43. [51]

    Completionformer: Depth completion with convolutions and vision transform- ers

    Youmin Zhang, Xianda Guo, Matteo Poggi, Zheng Zhu, Guan Huang, and Stefano Mattoccia. Completionformer: Depth completion with convolutions and vision transform- ers. In CVPR, pages 18527–18536, 2023. 3

    2023

  44. [52]

    Unleashing text-to-image diffu- sion models for visual perception

    Wenliang Zhao, Yongming Rao, Zuyan Liu, Benlin Liu, Jie Zhou, and Jiwen Lu. Unleashing text-to-image diffu- sion models for visual perception. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 5729–5739, 2023. 3

    2023

  45. [53]

    Semantic amodal segmentation

    Yan Zhu, Yuandong Tian, Dimitris Metaxas, and Piotr Doll´ar. Semantic amodal segmentation. In CVPR, pages 1464–1472, 2017. 3 11

    2017

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.