Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Controllable 3D Placement of Objects with Scene-Aware Diffusion Models

T0 review · 4 major / 6 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 The paper claims that a carefully designed visual map — a color-coded 3D bounding box projected onto the image plane — combined with coarse object masks is sufficient for high-quality object placement, and that this scheme achieves the…

desk verdict A genuinely novel conditioning map for 3D object placement, but the detector-based evaluation means the quantitative wins are plausible, not proven. read the letter →

arxiv 2506.21446 v1 pith:5Q3A6UNJ submitted 2025-06-26 cs.CV

Mohamed Omran , Dimitris Kalatzis , Jens Petersen , Amirhossein Habibian , Auke Wiggers This is my paper
classification cs.CV
keywords objectplacementdiffusionmodelsControlNet3DboundingboxconditioningimageinpaintingposefidelitynuScenesmonoculardetection
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 sets out to show that precise 3D placement of objects in existing images does not require elaborate masks, carefully engineered prompts, or learned coordinate tokens: a deliberately drawn visual map plus a coarse mask is enough. The map is a 3D bounding box rendered with two differently colored triangles per face, colored vertices, and a wireframe, then projected onto the image plane and fed through a ControlNet into a StableDiffusion inpainting model. On nuScenes driving scenes the method reports the lowest mean orientation error among the compared conditioning schemes in all three tasks studied — replacing an object, flipping it 180 degrees, and placing a car at a new location — while leaving the background unchanged by construction. The authors argue that orientation-accurate editing of this kind is what would let generative models produce pose-controlled training data for testing 3D detectors and other vision systems.

What carries the argument

The load-bearing object is the 'bbox3d map': a 3D bounding box rendered so that each of its six faces is split into two triangles of different colors, its eight vertices are colored, and its edges are drawn as a wireframe, then projected onto the image plane. The two triangles per face and the wireframe are what resolve the ambiguities the authors identify — which side of the object faces the camera, and whether the box is fully or only partially visible — ambiguities that single-color box maps and depth maps leave open. This map is fed through a pose encoder implemented as a ControlNet that conditions a StableDiffusion inpainting model, while the inpaint mask is a coarse, occlusion-aware mask obtained by projecting the box faces and subtracting the instance masks of occluding objects. The mechanism's defining property is that pose and shape are decoupled: the same projected box can drive edits that change shape, class, or orientation, whereas a depth map would freeze the shape and a coordinate token would hide the geometry.

What would settle it

Re-run the replacement benchmark with a second, independently trained 3D detector, or with human pose annotations on the edited crops, and check whether the orientation-error ordering persists — in particular whether the box map's mAOE of 0.121 stays ahead of the depth map's 0.133. If detector error correlates with the editing signal, the reported margins could be a measurement artifact rather than a true difference in placement accuracy.

Watch

Extended reading notes

Core claim

The paper's central claim is that a carefully designed visual map, combined with coarse object masks, is sufficient for high-quality 3D object placement. The conditioning signal is a color-coded cuboid: each of the six faces is filled with two triangles of distinct colors, the eight vertices are colored, and a wireframe is overlaid, so the projected image disambiguates which side faces the camera and whether the box is partially out of view; this map is processed by a ControlNet that conditions a StableDiffusion inpainting model on the desired pose and location. Measured with a pretrained monocular 3D detector on a filtered nuScenes validation set, the scheme achieves the best mean orientation error in all three settings — replacement (mAOE 0.121), flipped replacement (0.364), and placement into new locations (0.220) — and the fewest orientation flips in the two replacement settings, with FID close to the strongest baseline. The paper further shows that the same box map can be combined with an exemplar-based appearance encoder to preserve object identity while controlling pose, and that the two-triangle wireframe design beats simpler single-color box maps in the ablation.

Load-bearing premise

The evaluation assumes that the pretrained monocular 3D detector used to grade the edited images measures pose and location accurately and without bias toward any conditioning method, so that pre-filtering the test set to instances that detector already handles well does not skew the comparisons between methods.

Editorial extensions

If this is right

  • Because the method is built on an inpainting model, everything outside the coarse mask is left unchanged by construction, so a scene can be edited one object at a time at per-object resolution instead of regenerating the whole frame.
  • The same box-map conditioning supports edits the depth-based baselines cannot express: changing an object's class or shape, flipping its orientation 180 degrees, and filling in the newly exposed background.
  • Combining the box map with an exemplar appearance encoder places a specific vehicle at a specified location and orientation, cutting detector-measured orientation error from 0.456 to 0.181 mAOE relative to the exemplar-only baseline in replacement.
  • Explicit visual pose conditioning outperforms token-based pose encoders that must learn the projection from coordinate space to image space implicitly, even when those baselines receive more training iterations.
  • The conditioning map is cheap to synthesize from annotated 3D boxes, so the pipeline is directly usable for generating large, pose-controlled training sets that stress-test 3D detectors across orientations.

Reading between the lines

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

  • Testable extension: because the conditioning is purely geometric, the same projected-box map should transfer to other rigid objects such as pedestrians, furniture, or industrial parts, provided bounding boxes and camera calibration are available — the paper only demonstrates cars, trucks, and buses.
  • An implication the paper leaves implicit: the box map is a per-frame geometric anchor, so it could be handed to a video or multi-view generator to keep object pose consistent across frames; nothing in the method restricts it to single images.
  • The practical value of the wireframe map may lie less in average orientation error than in the cases depth cannot express — the narrow replacement margin over the depth baseline (0.121 vs 0.133 mAOE) suggests the larger wins are shape-changing edits and front/back-ambiguous objects such as buses.
  • If 3D detectors keep improving, the paper's measurement premise weakens: with near-perfect pose measurement the mAOE gaps between conditioning schemes would shrink, and realism and identity metrics would become the main differentiator between methods.
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

4 major / 6 minor

Summary. The paper proposes a conditioning signal for controllable 3D object placement in diffusion-based inpainting: a color-coded, wireframe 3D bounding box projected onto the image plane, fed through a ControlNet into a StableDiffusion inpainting model, with coarse masks derived from 3D bounding boxes. It evaluates on nuScenes with three tasks: object replacement, replacement with a 180-degree flip instruction, and placement in new locations, comparing against depth-based, coordinate-token, and other control baselines. The main quantitative claim is best orientation fidelity (lowest mAOE and flips) while preserving the background through an inpainting formulation. The paper also presents an exemplar-based variant that combines appearance control with pose control. The conclusion is that a carefully designed visual map, together with coarse object masks, is sufficient for high-quality object placement.

Significance. If the orientation-fidelity result holds, the method offers a simple and practical way to add 3D pose control to diffusion-based inpainting, decoupling shape from pose and enabling non-trivial edits such as class change or 180-degree rotation. The benchmark tasks are a useful template for evaluating 3D-aware image editing, and the paper is careful in ablating the conditioning design and comparing several control mechanisms. It also explicitly acknowledges limitations of 3D detectors in Section 5. However, because the quantitative evaluation depends on a monocular detector applied to edited, out-of-distribution images and lacks uncertainty quantification, the headline quantitative advantage is not yet firmly established.

major comments (4)
  1. [§4.1, Appendix A.1.1, Tables 1–3, Table 8] The evaluation protocol uses EPro-PnP-v2, a monocular 3D detector pretrained on real images, as the pose and location ground truth for edited images. The test set is filtered to instances that this detector already handles well on the original frame (yaw error below 3 degrees), but this filter does not control for detector error on edited images, which are out of distribution for the detector. The small quantitative margins in Tables 1–3 (e.g., 0.121 vs 0.133 in Table 1; 0.364 vs 0.387 in Table 2; 0.220 vs 0.230 in Table 3) could therefore reflect detector compatibility with each generator's output statistics rather than true orientation control, and Table 8 shows detector confidence varies by method (e.g., Ours 0.65, CN-Depth 0.67, GLIGEN 0.62 in replacement). The paper's own Section 5 notes that '3D detectors are not perfectly reliable,' making this the load-bearing assumption. I ask for a detector-agnostic check: report human judgments on a sample, evaluate on synthetic edits with known ground-truth poses, or at least show the relative ranking is stable across multiple independent detectors and detection confidence strata.
  2. [§4.1, Tables 1–3] The paper states that results are averaged over three random seeds, but Tables 1, 2, and 3 report only single mean values. No standard deviations, per-seed values, confidence intervals, or significance tests are given. Several comparisons are close (e.g., 0.121 vs 0.133 in Table 1; 0.220 vs 0.230 in Table 3; FID 9.36 vs 9.54 in Table 1), and without uncertainty it is impossible to tell whether the claimed advantage is real. Please report per-seed metrics or error bars and, where margins are small, a paired significance test.
  3. [Abstract, §4.2, Tables 1 and 3] The abstract and introduction emphasize 'precise location' as part of the contribution, but the reported translation errors do not support this. In the replacement task, Ours has mATE 1.390 versus CN-Depth 0.784 (Table 1); in placement, Ours has mATE 4.512 versus LooseControl 4.349 (Table 3). The method is consistently best on orientation (mAOE/flips) but not on location, and the oracle mATE is 0.353. The claim should be narrowed to orientation fidelity and the ability to specify location, rather than 'precise location' accuracy, unless additional evidence is provided.
  4. [§4.2, Appendix A.1.1] In the placement task, the evaluation compares detector output against the projected-box instruction on locations that contain no ground-truth object. The benchmark is therefore another detector-only measurement, and there is no evidence that the detector performs equally well for all methods on empty road regions. In addition, the comparison is not strictly controlled: GLIGEN and Neural Assets are trained for 500k/400k steps at frame level, while Ours uses 300k steps at crop level; the placement result could reflect training budget or crop context rather than the conditioning signal. Please report detector confidence and per-method failure rates on the placement task, and consider training all methods with matched budgets or at least discussing the effect of this mismatch.
minor comments (6)
  1. [§3] 'In this work choose to make the target pose explicit' is missing the subject 'we'.
  2. [Figure 1 caption] The caption says the method 'protect[s] foreground objects via their instance masks,' but §3 and Appendix A.1.2 describe occlusion-aware coarse masks derived from 3D bounding boxes; the caption should match the method description.
  3. [§4.2, Table 5] The 'six channels' ablation is not explained in the text; please specify how the six face channels are arranged and how 'faces' differs from 'mesh+wireframe' beyond the figure.
  4. [Appendix A.5] The appendix states that 'only 10% of all evaluation cases exhibit such errors in our case,' but Tables 1 and 2 report flip rates of 1.4% and 7.3% for Ours; the threshold and task to which the 10% statement refers should be defined.
  5. [References] References [12] and [13] are the same paper (Prompt-to-Prompt); one should be removed.
  6. [Tables 1 and 8] The FID values for the same methods differ across tables (e.g., LooseControl 9.54 in Table 1 but 14.37 in Table 8); either the settings differ or one table is outdated, and this should be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the box-map conditioning design is tested against external baselines and independent data, with no fitted quantity renamed as a prediction.

full rationale

The paper's central claim is that a projected, color-coded 3D bounding box used as ControlNet conditioning yields high pose fidelity for object placement. This is evaluated by comparing five externally defined baselines (SD-Inpaint, LooseControl, GLIGEN encodings, Neural Assets encodings, CN-Depth) that are all trained on the same nuScenes data with the same protocol; no parameter of the proposed method is fitted to the reported metrics. The evaluation uses a pretrained external detector (EPro-PnP-v2) to measure mAOE, mATE, and flips, and the detector is neither part of the generator nor optimized during training; the test-set filter for detector reliability is applied before any method is run, so it does not encode the proposed model's success. The ablation in Table 5 compares three conditioning-map designs on the same replacement task, but that is model selection on a fixed validation set, not a fitted parameter being reported as a prediction. There is one self-citation, [39], in Related Work supporting the peripheral statement that text-prompt insertion methods 'often alter the image in undesirable ways'; this is motivational, not load-bearing for the method, the ablations, or the pose-fidelity result. The authors' own Section 5 caveat that '3D detectors are not perfectly reliable' is a measurement-accuracy limitation, not a circularity. No equation or step defines the conditioning signal in terms of the measured outcome, so the derivation chain is self-contained.

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

The paper is empirical and introduces a conditioning representation rather than fitted constants or new physical entities. The only hand-chosen numeric inputs are evaluation thresholds in the benchmark, listed above. The load-bearing premises concern the validity of the measurement pipeline: detector-based pose metrics, filtered test selection, and crop-level FID. The authors explicitly acknowledge the need for accurate 3D annotations and calibration in Section 5.

free parameters (2)
  • Evaluation filtering thresholds = visibility >= 3; distance 4-40 m; min crop area 96x96 px; detector yaw error < 3 deg; crop expansion 1.5x
    Appendix A.1.1 and A.3.1 set these by hand to select the 5,000-instance benchmark. They determine the reported mAOE, mATE, and FID values, and the detector yaw filter removes hard cases, shaping all method comparisons.
  • Placement benchmark orientation grid = 8 orientations at 45-degree increments, 3 locations per frame, 200 frames
    Section 4.2 defines the placement task using this grid; the results in Table 3 depend on the chosen locations and orientation sampling.
assumptions (4)
  • domain assumption nuScenes 3D bounding box annotations and camera calibration are accurate enough to serve as both training targets and evaluation ground truth.
    Section 4.1 uses these annotations to extract training instances and create conditioning maps; Appendix A.1.2 projects 3D boxes to masks. If annotations are noisy, both pose conditioning and evaluation inherit that noise.
  • domain assumption The pretrained monocular 3D detector EPro-PnP-v2 measures pose and location on edited images without systematic bias across methods.
    Section 4.1 states the detector compares predictions on original and edited images; Appendix A.1.1 filters test instances based on detector accuracy. A biased detector would distort all mAOE and mATE comparisons.
  • domain assumption FID on square crops around objects is a valid realism measure despite real-background dominance and the use of a disjoint reference set.
    Section 4.1 computes FID on crops to avoid the background dominating; this assumes crop-level distribution matching captures edit realism.
  • domain assumption A StableDiffusion inpainting UNet and ControlNet, finetuned jointly for 300k steps, can learn the proposed conditioning without catastrophic forgetting or mode collapse.
    Section 3 describes the architecture and joint finetuning; this transferability is assumed rather than proven.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Controllable 3D Placement of Objects with Scene-Aware Diffusion Models." pith.science (2026). https://pith.science/paper/5Q3A6UNJ

@misc{pith2026250621446,
  author       = {Pith},
  title        = {Pith review of: Controllable 3D Placement of Objects with Scene-Aware Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5Q3A6UNJ}},
  note         = {Machine review of arXiv:2506.21446}
}
read the original abstract

Image editing approaches have become more powerful and flexible with the advent of powerful text-conditioned generative models. However, placing objects in an environment with a precise location and orientation still remains a challenge, as this typically requires carefully crafted inpainting masks or prompts. In this work, we show that a carefully designed visual map, combined with coarse object masks, is sufficient for high quality object placement. We design a conditioning signal that resolves ambiguities, while being flexible enough to allow for changing of shapes or object orientations. By building on an inpainting model, we leave the background intact by design, in contrast to methods that model objects and background jointly. We demonstrate the effectiveness of our method in the automotive setting, where we compare different conditioning signals in novel object placement tasks. These tasks are designed to measure edit quality not only in terms of appearance, but also in terms of pose and location accuracy, including cases that require non-trivial shape changes. Lastly, we show that fine location control can be combined with appearance control to place existing objects in precise locations in a scene.

Figures

Figures reproduced from arXiv: 2506.21446 by the authors.

Figure 1
Figure 1. Overview of the use case and method. We specify a [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Various conditioning signals for 2.5D and 3D location control based on depth maps or 3D bounding boxes. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Object replacement examples with flipped orientation instruction for various scenes, vehicles and occlusion levels. Shown here [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: Samples from our method for the object placement task, showing (zoomed in) editing results for 8 target orientations for three [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: Example generations from our method and baseline methods, changing class but preserving orientation. Only GLIGEN [ [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Samples from our method with exemplar-prompt and specific pose instruction. Top: original object frames. Bottom: samples with [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Detailed inpainting masks (8b) that fit the original object exactly do not allow for replacement with non-trivial shape change. Coarse masks derived from 3D bounding boxes on the other hand (8c) require inpainting models to fill in occluding foreground objects besides …
Figure 9
Figure 9. Figure 9: Overview of the exemplar-conditioning pipeline. The appearance encoder [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 10
Figure 10. Figure 10: Example generations from all models in the standard object editing evaluation setting. [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: Example generations from all models in the object editing evaluation setting with flipped pose instruction. [PITH_FULL_IMAGE:figures/full_fig_p015_11.png]
Figure 12
Figure 12. Figure 12: Placing specific vehicles in specific locations. The left column shows the exemplar, the other columns show a the result of [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]
Figure 13
Figure 13. Figure 13: Placing specific vehicles in specific locations. The left column shows the exemplar, the other columns show a the result of [PITH_FULL_IMAGE:figures/full_fig_p016_13.png]
Figure 14
Figure 14. Figure 14: Placing specific vehicles in specific locations. The left column shows the exemplar, the other columns show a the result of [PITH_FULL_IMAGE:figures/full_fig_p017_14.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. BoxCtrl: 3D-Aware Visual Prompting for Geometric Image Editing

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    BoxCtrl introduces colored 3D bounding boxes as visual prompts for geometric image editing, trained first on synthetic data via supervised fine-tuning then refined with reinforcement learning on real data.

Reference graph

Works this paper leans on

40 extracted references · 25 canonical work pages · cited by 1 Pith paper

  1. [1]

    DiffUHaul: A training-free method for object dragging in images

    Omri Avrahami, Rinon Gal, Gal Chechik, Ohad Fried, Dani Lischinski, Arash Vahdat, and Weili Nie. DiffUHaul: A training-free method for object dragging in images. In SIG- GRAPH Asia 2024 Conference Papers, pages 1–12, 2024. 2

    work page 2024

  2. [2]

    LooseC- ontrol: Lifting controlnet for generalized depth conditioning

    Shariq Farooq Bhat, Niloy Mitra, and Peter Wonka. LooseC- ontrol: Lifting controlnet for generalized depth conditioning. In ACM SIGGRAPH 2024 Conference Papers , pages 1–11,

    work page 2024

  3. [3]

    In- structpix2pix: Learning to follow image editing instructions

    Tim Brooks, Aleksander Holynski, and Alexei A Efros. In- structpix2pix: Learning to follow image editing instructions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18392–18402, 2023. 2

    2023

  4. [4]

    nuscenes: A multi- modal dataset for autonomous driving

    Holger Caesar, Varun Bankiti, Alex H Lang, Sourabh V ora, Venice Erin Liong, Qiang Xu, Anush Krishnan, Yu Pan, Gi- ancarlo Baldan, and Oscar Beijbom. nuscenes: A multi- modal dataset for autonomous driving. In Proceedings of the IEEE conference on Computer Vision and Pattern Recogni- tion, 2020. 3, 4

    work page 2020

  5. [5]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. In IEEE International Conference on Computer Vision , pages 9650–9660, 2021. 6, 3

    work page 2021

  6. [6]

    Epro-PnP: Generalized end-to-end prob- abilistic perspective-n-points for monocular object pose es- timation

    Hansheng Chen, Pichao Wang, Fan Wang, Wei Tian, Lu Xiong, and Hao Li. Epro-PnP: Generalized end-to-end prob- abilistic perspective-n-points for monocular object pose es- timation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2781–2790,

    work page

  7. [7]

    Geodiffusion: Text- prompted geometric control for object detection data genera- tion

    Kai Chen, Enze Xie, Zhe Chen, Yibo Wang, Lanqing Hong, Zhenguo Li, and Dit-Yan Yeung. Geodiffusion: Text- prompted geometric control for object detection data genera- tion. International Conference on Learning Representations,

    work page

  8. [8]

    Text-to-3d using gaussian splatting

    Zilong Chen, Feng Wang, Yikai Wang, and Huaping Liu. Text-to-3d using gaussian splatting. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21401–21412, 2024. 2

    work page 2024

Show all 40 references
  1. [9]

    Carla: An open urban driv- ing simulator

    Alexey Dosovitskiy, German Ros, Felipe Codevilla, Anto- nio Lopez, and Vladlen Koltun. Carla: An open urban driv- ing simulator. In Conference on robot learning, pages 1–16. PMLR, 2017. 1

    2017

  2. [10]

    Build-a-scene: Interactive 3d layout control for diffusion-based image gen- eration

    Abdelrahman Eldesokey and Peter Wonka. Build-a-scene: Interactive 3d layout control for diffusion-based image gen- eration. arXiv preprint arXiv:2408.14819, 2024. 2

    2024 arXiv

  3. [11]

    Magicdrive: Street view generation with diverse 3d geometry control

    Ruiyuan Gao, Kai Chen, Enze Xie, Lanqing Hong, Zhenguo Li, Dit-Yan Yeung, and Qiang Xu. Magicdrive: Street view generation with diverse 3d geometry control. International Conference on Learning Representations, 2024. 1, 2, 3

    2024

  4. [13]

    Prompt-to-prompt im- age editing with cross attention control

    Amir Hertz, Ron Mokady, Jay Tenenbaum, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Prompt-to-prompt im- age editing with cross attention control. arXiv preprint arXiv:2208.01626, 2022. 2

    2022 arXiv

  5. [14]

    Gans trained by a two time-scale update rule converge to a local nash equilib- rium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilib- rium. In Neural Information Processing Systems, 2017. 4

    2017

  6. [15]

    Gaia-1: A generative world model for au- tonomous driving

    Anthony Hu, Lloyd Russell, Hudson Yeo, Zak Murez, George Fedoseev, Alex Kendall, Jamie Shotton, and Gian- luca Corrado. Gaia-1: A generative world model for au- tonomous driving. arXiv preprint arXiv:2309.17080, 2023. 1

    2023 arXiv

  7. [16]

    An edit friendly ddpm noise space: Inversion and manipulations

    Inbar Huberman-Spiegelglas, Vladimir Kulikov, and Tomer Michaeli. An edit friendly ddpm noise space: Inversion and manipulations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12469– 12478, 2024. 2

    2024

  8. [17]

    Dense text-to-image generation with attention modulation

    Yunji Kim, Jiyoung Lee, Jin-Hwa Kim, Jung-Woo Ha, and Jun-Yan Zhu. Dense text-to-image generation with attention modulation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 7701–7711, 2023. 1

    2023

  9. [18]

    Berg, Wan-Yen Lo, Piotr Dollar, and Ross Girshick

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C. Berg, Wan-Yen Lo, Piotr Dollar, and Ross Girshick. Segment anything. In Proceedings of the IEEE/CVF International Conference on Computer Visio...

    2023

  10. [19]

    GLIGEN: open-set grounded text-to-image generation

    Yuheng Li, Haotian Liu, Qingyang Wu, Fangzhou Mu, Jianwei Yang, Jianfeng Gao, Chunyuan Li, and Yong Jae Lee. GLIGEN: open-set grounded text-to-image generation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17- 24, 20...

    2023

  11. [20]

    Magic3d: High-resolution text-to-3d content creation

    Chen-Hsuan Lin, Jun Gao, Luming Tang, Towaki Takikawa, Xiaohui Zeng, Xun Huang, Karsten Kreis, Sanja Fidler, Ming-Yu Liu, and Tsung-Yi Lin. Magic3d: High-resolution text-to-3d content creation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognitio...

    2023

  12. [21]

    One-2-3-45: Any single image to 3d mesh in 45 seconds without per-shape optimiza- tion

    Minghua Liu, Chao Xu, Haian Jin, Linghao Chen, Mukund Varma T, Zexiang Xu, and Hao Su. One-2-3-45: Any single image to 3d mesh in 45 seconds without per-shape optimiza- tion. Advances in Neural Information Processing Systems , 36, 2024. 2

    2024

  13. [22]

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

    Ruoshi Liu, Rundi Wu, Basile Van Hoorick, Pavel Tok- makov, Sergey Zakharov, and Carl V ondrick. Zero-1-to- 3: Zero-shot one image to 3d object. In Proceedings of the IEEE/CVF international conference on computer vision, pages 9298–9309, 2023. 1, 2, 4

    2023

  14. [23]

    Object 3dit: Language-guided 3d-aware image editing

    Oscar Michel, Anand Bhattad, Eli VanderBilt, Ranjay Kr- ishna, Aniruddha Kembhavi, and Tanmay Gupta. Object 3dit: Language-guided 3d-aware image editing. Advances in Neural Information Processing Systems, 36, 2024. 4

    2024

  15. [24]

    T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models

    Chong Mou, Xintao Wang, Liangbin Xie, Yanze Wu, Jian Zhang, Zhongang Qi, and Ying Shan. T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 4296–4304, 2024. 1

    2024

  16. [25]

    Dreamfusion: Text-to-3d using 2d diffusion

    Ben Poole, Ajay Jain, Jonathan T Barron, and Ben Milden- hall. Dreamfusion: Text-to-3d using 2d diffusion. arXiv preprint arXiv:2209.14988, 2022. 1, 2

    2022 arXiv

  17. [26]

    Magic123: One image to high-quality 3d object generation using both 2d and 3d diffusion priors.arXiv preprint arXiv:2306.17843,

    Guocheng Qian, Jinjie Mai, Abdullah Hamdi, Jian Ren, Aliaksandr Siarohin, Bing Li, Hsin-Ying Lee, Ivan Sko- rokhodov, Peter Wonka, Sergey Tulyakov, et al. Magic123: One image to high-quality 3d object generation using both 2d and 3d diffusion priors.arXiv preprint arXiv:2306.17843,

    arXiv

  18. [27]

    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 Proceedings of the IEEE conference on Computer Vision and Pattern Recogni- tion, 2022. 1, 2, 3, 5, 7, 4

    2022

  19. [28]

    Airsim: High-fidelity visual and physical simulation for autonomous vehicles

    Shital Shah, Debadeepta Dey, Chris Lovett, and Ashish Kapoor. Airsim: High-fidelity visual and physical simulation for autonomous vehicles. In Field and Service Robotics: Re- sults of the 11th International Conference , pages 621–635. Springer, 2018. 1

    2018

  20. [29]

    Object- stitch: Object compositing with diffusion model

    Yizhi Song, Zhifei Zhang, Zhe Lin, Scott Cohen, Brian Price, Jianming Zhang, Soo Ye Kim, and Daniel Aliaga. Object- stitch: Object compositing with diffusion model. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18310–18319, 2023. 2

    2023

  21. [30]

    Make-it-3d: High-fidelity 3d creation from a single image with diffusion prior

    Junshu Tang, Tengfei Wang, Bo Zhang, Ting Zhang, Ran Yi, Lizhuang Ma, and Dong Chen. Make-it-3d: High-fidelity 3d creation from a single image with diffusion prior. In Pro- ceedings of the IEEE/CVF international conference on com- puter vision, pages 22819–22829, 2023. 2

    2023

  22. [31]

    Plug-and-play diffusion features for text-driven image-to-image translation

    Narek Tumanyan, Michal Geyer, Shai Bagon, and Tali Dekel. Plug-and-play diffusion features for text-driven image-to-image translation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 1921–1930, 2023. 2

    1921

  23. [32]

    Paint by inpaint: Learning to add image objects by removing them first

    Navve Wasserman, Noam Rotstein, Roy Ganz, and Ron Kimmel. Paint by inpaint: Learning to add image objects by removing them first. arXiv preprint arXiv:2404.18212 ,

    arXiv

  24. [33]

    Objectdrop: Bootstrap- ping counterfactuals for photorealistic object removal and in- sertion

    Daniel Winter, Matan Cohen, Shlomi Fruchter, Yael Pritch, Alex Rav-Acha, and Yedid Hoshen. Objectdrop: Bootstrap- ping counterfactuals for photorealistic object removal and in- sertion. arXiv preprint arXiv:2403.18818, 2024. 2

    2024 arXiv

  25. [34]

    Neural assets: 3d-aware multi-object scene synthesis with image diffusion models

    Ziyi Wu, Yulia Rubanova, Rishabh Kabra, Drew A Hud- son, Igor Gilitschenski, Yusuf Aytar, Sjoerd van Steenkiste, Kelsey R Allen, and Thomas Kipf. Neural assets: 3d-aware multi-object scene synthesis with image diffusion models. arXiv preprint arXiv:2406.09292, 2024. 1, 2, 3, 4...

    2024 arXiv

  26. [35]

    Paint by example: Exemplar-based image editing with diffusion mod- els

    Binxin Yang, Shuyang Gu, Bo Zhang, Ting Zhang, Xuejin Chen, Xiaoyan Sun, Dong Chen, and Fang Wen. Paint by example: Exemplar-based image editing with diffusion mod- els. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 18381–18391,

  27. [36]

    Depth anything: Unleashing the power of large-scale unlabeled data

    Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, and Hengshuang Zhao. Depth anything: Unleashing the power of large-scale unlabeled data. In IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, WA, USA, June 16-22, 2024 , pages 1...

    2024

  28. [37]

    Unisim: A neural closed-loop sensor simulator

    Ze Yang, Yun Chen, Jingkang Wang, Sivabalan Mani- vasagam, Wei-Chiu Ma, Anqi Joyce Yang, and Raquel Ur- tasun. Unisim: A neural closed-loop sensor simulator. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 1389–1399, 2023. 1

    2023

  29. [38]

    Gaussian- dreamer: Fast generation from text to 3d gaussian splatting with point cloud priors

    Taoran Yi, Jiemin Fang, Guanjun Wu, Lingxi Xie, Xiaopeng Zhang, Wenyu Liu, Qi Tian, and Xinggang Wang. Gaussian- dreamer: Fast generation from text to 3d gaussian splatting with point cloud priors. arXiv preprint arXiv:2310.08529 ,

    arXiv

  30. [39]

    Generative lo- cation modeling for spatially aware object insertion

    Jooyeol Yun, Davide Abati, Mohamed Omran, Jaegul Choo, Amirhossein Habibian, and Auke Wiggers. Generative lo- cation modeling for spatially aware object insertion. arXiv preprint arXiv:2410.13564, 2024. 2

    2024

  31. [40]

    Adding conditional control to text-to-image diffusion models, 2023

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models, 2023. 1, 2, 3, 5, 7, 4

    2023

  32. [41]

    Layoutdiffusion: Controllable diffu- sion model for layout-to-image generation

    Guangcong Zheng, Xianpan Zhou, Xuewei Li, Zhongang Qi, Ying Shan, and Xi Li. Layoutdiffusion: Controllable diffu- sion model for layout-to-image generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 22490–22499, 2023. 1 Contr...

    2023

Pith tools

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