Pith. sign in

REVIEW 2 major objections 6 minor 3 cited by

Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real Vehicles

T0 review · 2 major / 6 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 Retuning a synthetic-view diffusion model with four practices yields photorealistic 360-degree views of real vehicles from one image.

desk verdict A practical finetuning recipe that delivers real gains on Waymo vehicles; the symmetry prior is the main unresolved risk, but the paper is worth a serious referee. read the letter →

arxiv 2412.14494 v1 pith:ZUIVIIFG submitted 2024-12-19 cs.CV

Chuang Lin , Bingbing Zhuang , Shanlin Sun , Ziyu Jiang , Jianfei Cai , Manmohan Chandraker This is my paper · ORCID
classification cs.CV
keywords novelviewsynthesisdiffusionmodelsposeconditioningautonomousdrivingdomainfinetuningvehicle3Dassetsocclusion-awaretrainingsymmetricprior
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 claims that a pose-conditioned diffusion model trained on synthetic 3D objects can be finetuned, cheaply, to generate photorealistic novel views of real vehicles from a single image, and that four specific practices make the transfer work. The authors start from a pretrained model that already knows generic 3D shape and image appearance, and adapt it to driving data rather than training a vehicle-specific model from scratch. They report a 68.8% reduction in FID over prior vehicle-specific methods, and show the generated 360-degree views feed a downstream model that reconstructs a drivable 3D asset and inserts it into a simulator. If the claim is right, autonomous-driving simulation can harvest vehicle assets from ordinary road logs without costly manual 3D modeling.

What carries the argument

The load-bearing object is the pose-conditioned latent diffusion model itself: a U-Net denoiser conditioned on a source image and on the relative camera pose, expressed both as a global rotation-translation pair and as per-pixel ray embeddings that encode each pixel's viewing ray (the Free3D design). Against this backbone the paper applies four interventions, each aimed at one measurable mismatch between synthetic renders and real driving frames: orbital virtual camera rotation via a rotational homography, fixed-focal-length object cropping, occlusion-masked noise-prediction loss in latent space, and symmetric-pair training through horizontal flipping with negated azimuth. The identity that carries the argument is the pose-conditioned denoising objective, modified to mask occluded latent elements.

What would settle it

Compute the 60-180 degree FID and LPIPS on a held-out subset of vehicles with visibly asymmetric details, such as delivery van lettering, one-side damage, or motorcycle sidecars, after training with and without the symmetric-pair augmentation; if the gap between the two training regimes narrows or reverses on that subset, the symmetry prior is the fragile load-bearing assumption.

Watch

Extended reading notes

Core claim

Drive-1-to-3 is the claim that the gap between synthetic rendering and real road imagery is bridgeable by aligning real images to the pretrained model's canonical pose space rather than retraining. Concretely: each real vehicle crop is virtually rotated, keeping the camera center fixed, so the image looks as if captured from an orbital camera and the pose can be expressed by elevation, azimuth, and distance; crops are made at a fixed focal length so object scale varies but perspective distortion stays stable; occlusion masks are downsampled into latent space and the noise-prediction loss ignores those regions; and each source image is paired with its horizontally flipped twin with negated azimuth to create large viewpoint differences during training. The authors argue that these choices let the network keep the generic priors of the pretrained diffusion model while learning vehicle-specific appearance from real driving data, and report that the combination yields photorealistic 360-degree views, with large gains under 60-180 degree azimuth changes.

Load-bearing premise

The method assumes vehicles are bilaterally symmetric enough that horizontally flipping the source image and negating its azimuth produces a valid training pair, so features like side lettering, asymmetric mirrors, exhaust layout, or one-sided damage must not systematically break the learned pose mapping.

Editorial extensions

If this is right

  • A single photo of a vehicle from an onboard camera can be turned into a 360-degree view set that is visually consistent enough for downstream multi-view 3D reconstruction.
  • Vehicle assets for simulation can be harvested automatically from ordinary driving logs instead of being built by hand, at a finetuning cost of roughly ten GPU-hours.
  • For viewpoint changes of 60-180 degrees in azimuth, the method maintains much lower FID and LPIPS than category-specific NeRF and GAN baselines, so large-angle synthesis is not limited by the mostly-forward motion of real driving data.
  • The same pretrained diffusion backbone, after finetuning, generalizes to other real vehicle datasets such as NuScenes and DVM-Car without per-dataset retraining.
  • Rendering 360-degree views into a large multi-view Gaussian reconstruction model yields a 3D asset that can be inserted into a neural driving simulator with plausible shadows.

Reading between the lines

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

  • The symmetry prior is the most fragile component: vehicles with asymmetric markings, one-sided damage, or non-symmetric accessories should degrade the 60-180 degree gains, a prediction that could be tested by training on a filtered asymmetric subset.
  • The fixed-focal-length finding suggests a general design rule for finetuning pose-conditioned diffusion models to any object category captured in the wild: preserve the pretrained model's camera intrinsics and let the network absorb scale variation, rather than normalizing scale by changing field of view.
  • Occlusion-aware latent masking may carry over to other latent diffusion tasks such as inpainting or editing of partially occluded objects, since the paper confirms the latent-space mask transfers to image space.
  • The success of orbital virtual rotation implies any real image with known camera intrinsics and object depth could be reposed into the pretrained model's canonical manifold, potentially extending the approach beyond vehicles to pedestrians, cyclists, or general street furniture, provided a symmetry or multi-view prior is available.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 6 minor

Summary. This paper proposes Drive-1-to-3, a finetuning recipe for adapting a pretrained pose-conditioned diffusion model (Free3D) to novel view synthesis of real vehicles from a single onboard camera image. The method consolidates four practices: virtual rotation of real images into an orbital camera pose, object-centric cropping with constant focal length and varying object scale, occlusion-aware loss in latent space, and strong symmetric-pair training that horizontally flips source images to enlarge viewpoint variation. Experiments on the Waymo Open Dataset show substantial improvements in FID, LPIPS, and other metrics over AutoRF, DisCoScene, and the un-finetuned Free3D baseline, and the paper demonstrates downstream use in LGM-based 3D reconstruction and virtual object insertion into driving simulations.

Significance. If the reported results are reproducible, this is a useful empirical contribution to domain transfer for diffusion-based novel view synthesis. The four practices are clearly motivated, individually ablated, and the combined system achieves a large FID reduction (68.8% versus DisCoScene) with modest training cost (10 hours on a single A6000 GPU). The paper also provides qualitative evidence for cross-dataset generalization and shows a practical downstream application. The main weaknesses are that the symmetric-prior assumption is load-bearing and insufficiently examined, and that the FID evaluation protocol is underspecified, which affects the strength and reproducibility of the central claim.

major comments (2)
  1. [§3.4, Table 2] The symmetric-prior training is load-bearing: comparing rows (d) and (g) in Table 2 shows large gains from adding the symmetry prior, especially for large viewpoint changes (60–180° PSNR 15.909→17.838, LPIPS 0.198→0.130, and FID 7.09→4.18). However, the method constructs training pairs by horizontally flipping the source image and negating its azimuth, which is physically valid only for perfectly bilaterally symmetric vehicles. Real vehicles contain asymmetric details such as side lettering, fuel filler placement, asymmetric exhausts, antennas, and one-sided damage. The manuscript does not analyze how often the model mirror-reverses such details, does not report performance broken down by vehicle asymmetry, and does not discuss whether the strong guidance causes the model to hallucinate symmetric textures on asymmetric instances. Since the downstream LGM reconstruction and object insertion would propagate mirrored textures into the final asset, this is a correctness risk in the target application. Please add a quantitative analysis (e.g., measuring left–right consistency of predictions on a subset annotated for asymmetry, or reporting failure cases) and discuss possible mitigations.
  2. [§4.1] The FID evaluation protocol is underspecified. The text says that 360° views are rendered at 90-degree intervals and evaluated using dataset-level Fréchet Inception Distance 'in the absence of ground truth,' but it does not state which real images form the reference distribution for the FID computation. This matters directly for the abstract's headline 68.8% FID reduction (4.18 vs 13.40 in Table 1). Please specify exactly what the reference set is (e.g., all real vehicle crops from the Waymo training split, a held-out set of real target views, or another distribution), how the four rendered views per object are aggregated, and the image size and pre-processing used for FID.
minor comments (6)
  1. [§1, §2, §3.2] There are several typos: 'serval' should be 'several' (§1), 'ambiguious' should be 'ambiguous' (§2), and 'Zero-1-2-3' should be 'Zero-1-to-3' (§3.2 and supplementary).
  2. [Table 1] The table formatting is confusing: FID values and angle intervals are concatenated, e.g., '21.8530-30' should read as two separate entries (FID 21.85 and angle range 0–30). Please reformat all rows so that FID and angle ranges are clearly separated.
  3. [Figure 6 caption] The caption refers to 'MVD-Car' but the dataset is DVM-Car; correct the typo.
  4. [§3.3] The text says 'we first downsample the target view occlusion mask m' but the motivation for occlusion handling is occlusions in the source view (see Introduction). Please clarify whether the mask is computed from the source image, the target image, or both, and how it is used in the loss.
  5. [§3.4] The comparison between weak and strong symmetry guidance is demonstrated only qualitatively in Figure 5(b). Given that the paper recommends strong guidance as a good practice, a quantitative ablation in Table 2 would allow readers to assess the magnitude of this design choice.
  6. [Supplementary] In the supplementary text, the sentence 'Additional results on NuScenes and DVM-Car dataset are shown in are shown in Fig. 15' contains a duplicated phrase; please fix.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's headline results are empirical finetuning comparisons against external baselines on held-out real vehicle views.

full rationale

The central claim is that four finetuning practices enable a Free3D pretrained diffusion model to synthesize photorealistic novel views of real vehicles. This is an empirical finetuning study, not a derivation in which a predicted quantity is defined in terms of its own inputs. The headline 68.8% FID reduction is a direct comparison of dataset-level FID scores computed on rendered 360-degree views: Free3D-style baselines are externally trained, and the FID is measured without any fitting of the metric to the method's outputs. The quantitative comparisons in Table 1 and ablations in Table 2 use real ground-truth target views from Waymo, with occlusion masking applied consistently to account for the absence of occluders in predictions; this is an evaluation adjustment, not a circular redefinition. The backbone Free3D is an external pretrained model, not a result of the present authors, and no load-bearing argument reduces to a self-citation. The symmetric prior in Section 3.4 is an explicit modeling assumption about the vehicle category; it is not claimed to be derived from the method itself, and its effectiveness is measured empirically by comparing configurations with and without it. The paper does not fit a parameter to a subset of data and then predict that same subset, nor does it rename a known empirical pattern as a derivation. The correctness risk around asymmetric vehicles is a domain-validity concern, not circularity. Overall, the derivation chain is self-contained in the sense that the reported gains are not forced by construction or by self-referential definitions.

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

The central claim rests on hand-chosen design choices, notably fixed-FOV cropping, 3-degree azimuth sampling, the occlusion category list, and strong symmetric pairing, plus domain assumptions about the pretrained prior, rotational homographies, latent-space mask transfer, and vehicle symmetry. No new physical entities are introduced.

free parameters (4)
  • Training view sampling stride = 3 degrees in azimuth
    Hand-selected threshold for collecting source-target pairs from Waymo driving sequences; controls pose diversity and training set size (Section 4.1).
  • Object-crop field of view = Not reported; fixed to the Objaverse rendering field of view
    Hand-chosen strategy to keep focal length fixed while object scale varies; the paper argues this is easier for the pretrained encoder (Section 3.2, Figure 3).
  • Occlusion mask category set = 17 Waymo panoptic categories, including vehicles, poles, pedestrians, and vegetation
    Hand-specified list of plausible occluders used to build training masks and to exclude pixels from the loss (Supplementary, Section 6).
  • Symmetric guidance strength = Strong guidance, paired symmetric images in the same batch
    Selected by ablation over weak 50 percent flip augmentation; strong guidance yields better large-viewpoint performance (Section 3.4, Table 2).
assumptions (5)
  • domain assumption Waymo 3D object boxes provide accurate absolute poses for all training vehicles.
    Used to compute orbital virtual rotations and relative pose conditioning; pose errors would misalign source and target views (Sections 3.2, 4.1).
  • standard math A rotation of the camera about its center is exactly represented by a rotational homography.
    Justifies the virtual camera rotation step; relies on standard multi-view geometry from Hartley and Zisserman (Section 3.2).
  • domain assumption Downsampling a binary image-space occlusion mask to latent resolution and masking the noise-prediction loss effectively masks image-space occlusions.
    The paper states there is no exact pixel-to-latent correspondence and relies on Blended Latent Diffusion evidence; it is empirical rather than proven (Section 3.3).
  • domain assumption Vehicles are bilaterally symmetric enough that horizontal flipping with azimuth negation produces valid training views.
    Used to expand the viewpoint change span in training data (Section 3.4); false in detail for many real vehicles.
  • domain assumption The Free3D and Zero-1-to-3 pose prior learned on synthetic Objaverse renders transfers to real vehicle images after virtual rotation and finetuning.
    This is the central premise of the finetuning approach; it is supported only empirically through ablations (Section 4.3).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real Vehicles." pith.science (2026). https://pith.science/paper/ZUIVIIFG

@misc{pith2026241214494,
  author       = {Pith},
  title        = {Pith review of: Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real Vehicles},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZUIVIIFG}},
  note         = {Machine review of arXiv:2412.14494}
}
abstract

The recent advent of large-scale 3D data, e.g. Objaverse, has led to impressive progress in training pose-conditioned diffusion models for novel view synthesis. However, due to the synthetic nature of such 3D data, their performance drops significantly when applied to real-world images. This paper consolidates a set of good practices to finetune large pretrained models for a real-world task -- harvesting vehicle assets for autonomous driving applications. To this end, we delve into the discrepancies between the synthetic data and real driving data, then develop several strategies to account for them properly. Specifically, we start with a virtual camera rotation of real images to ensure geometric alignment with synthetic data and consistency with the pose manifold defined by pretrained models. We also identify important design choices in object-centric data curation to account for varying object distances in real driving scenes -- learn across varying object scales with fixed camera focal length. Further, we perform occlusion-aware training in latent spaces to account for ubiquitous occlusions in real data, and handle large viewpoint changes by leveraging a symmetric prior. Our insights lead to effective finetuning that results in a $68.8\%$ reduction in FID for novel view synthesis over prior arts.

Figures

Figures reproduced from arXiv: 2412.14494 by the authors.

Figure 1
Figure 1. Comparisons between the pretrained Free3D [ [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall pipeline of Drive-1-to-3. First, it processes a single vehicle image from on-board cameras, virtually rotating it to a shared orbital pose. The object-centric image cropped with a constant focal length is fed to a pose-conditioned diffusion model, which performs occlusion-aware training in latent space with a symmetric prior. multiview view synthesis machine. 3D learning for real-world vehicles. Single-i… view at source ↗
Figure 3
Figure 3. Illustration of two strategies in object-centric image [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of our method with AutoRF [ [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Qualitative results showing benefits of (a) occlusion handling and (b) symmetric prior. Inputs Novel View Synthesis (a) Night condition (b) Unusual vehicles Inputs (c) Reflect surface Inputs Novel View Synthesis (a) Night condition (b) Unusual vehicles Inputs (c) Refle…
Figure 7
Figure 7. Figure 7: Qualitative comparison of varying object scales and varying focal lengths. The example in the blue dashed box shows a significant [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: LGM 3D reconstruction comparison with Drive-1-to-3 versus its default MV [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 10
Figure 10. Figure 10: Illustration of unaligned azimuth in Objaverse vs. the [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 11
Figure 11. Figure 11: Example occlusion mask, where white pixels indicate [PITH_FULL_IMAGE:figures/full_fig_p011_11.png]
Figure 13
Figure 13. Figure 13: Additional results showing benefits of (a) occlusion handling and (b) symmetric prior. [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]
Figure 14
Figure 14. Figure 14: Qualitative comparison of varying object scales and varying focal lengths. The examples in the blue dashed boxes show a [PITH_FULL_IMAGE:figures/full_fig_p012_14.png]
Figure 15
Figure 15. Figure 15: Additional results on NuScenes and DVM-Car dataset. [PITH_FULL_IMAGE:figures/full_fig_p013_15.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. Reference-Guided Diffusion Inpainting For Multimodal Counterfactual Generation

    cs.CV 2025-07 conditional novelty 6.0 of 10

    A single reference image guides a diffusion model to insert coherent objects into camera-plus-lidar driving scenes and to insert mammographic anomalies into new scans.

  2. MObI: Multimodal Object Inpainting Using Diffusion Models

    cs.CV 2025-01 conditional novelty 6.0 of 10

    MObI jointly inpaints camera and lidar views of driving scenes, inserting objects from a single reference image at a user-specified 3D bounding box.

  3. ArbiViewGen: Controllable Arbitrary Viewpoint Camera Data Generation for Autonomous Driving via Stable Diffusion Models

    cs.CV 2025-08 conditional novelty 5.0 of 10

    ArbiViewGen generates arbitrary-viewpoint driving camera images by stitching the six input views into pseudo-target views and training a Stable Diffusion model to reconstruct the original views, enabling self-supervis...

Reference graph

Works this paper leans on

53 extracted references · 29 canonical work pages · cited by 3 Pith papers

  1. [1]

    Blended latent diffusion

    Omri Avrahami, Ohad Fried, and Dani Lischinski. Blended latent diffusion. ACM Transactions on Graphics (TOG), 42 (4):1–11, 2023. 5

    work page 2023

  2. [2]

    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 CVPR, 2020. 6, 8

    work page 2020

  3. [3]

    Objaverse: A universe of annotated 3d objects

    Matt Deitke, Dustin Schwenk, Jordi Salvador, Luca Weihs, Oscar Michel, Eli VanderBilt, Ludwig Schmidt, Kiana Ehsani, Aniruddha Kembhavi, and Ali Farhadi. Objaverse: A universe of annotated 3d objects. In CVPR, 2023. 1, 2, 3, 5

    work page 2023

  4. [4]

    Objaverse-xl: A universe of 10m+ 3d objects

    Matt Deitke, Ruoshi Liu, Matthew Wallingford, Huong Ngo, Oscar Michel, Aditya Kusupati, Alan Fan, Christian Laforte, Vikram V oleti, Samir Yitzhak Gadre, et al. Objaverse-xl: A universe of 10m+ 3d objects. NeurIPS, 2024. 6

    work page 2024

  5. [5]

    3drealcar: An in-the-wild rgb- d car dataset with 360-degree views

    Xiaobiao Du, Haiyang Sun, Shuyun Wang, Zhuojie Wu, Hongwei Sheng, Jiaying Ying, Ming Lu, Tianqing Zhu, Kun Zhan, and Xin Yu. 3drealcar: An in-the-wild rgb- d car dataset with 360-degree views. arXiv preprint arXiv:2406.04875, 2024. 3

    arXiv 2024

  6. [6]

    Streetsurf: Extending multi-view im- plicit surface reconstruction to street views

    Jianfei Guo, Nianchen Deng, Xinyang Li, Yeqi Bai, Bo- tian Shi, Chiyu Wang, Chenjing Ding, Dongliang Wang, and Yikang Li. Streetsurf: Extending multi-view im- plicit surface reconstruction to street views. arXiv preprint arXiv:2306.04988, 2023. 1

    arXiv 2023

  7. [7]

    Vfusion3d: Learning scalable 3d generative models from video diffusion models

    Junlin Han, Filippos Kokkinos, and Philip Torr. Vfusion3d: Learning scalable 3d generative models from video diffusion models. In ECCV, 2024. 2

    work page 2024

  8. [8]

    Multiple view ge- ometry in computer vision

    Richard Hartley and Andrew Zisserman. Multiple view ge- ometry in computer vision . Cambridge university press,

Show all 53 references
  1. [9]

    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. NIPS, 2017. 6

    2017

  2. [10]

    Lrm: Large reconstruction model for single image to 3d

    Yicong Hong, Kai Zhang, Jiuxiang Gu, Sai Bi, Yang Zhou, Difan Liu, Feng Liu, Kalyan Sunkavalli, Trung Bui, and Hao Tan. Lrm: Large reconstruction model for single image to 3d. arXiv preprint arXiv:2311.04400, 2023. 2

    2023 arXiv

  3. [11]

    Dvm-car: A large-scale automotive dataset for visual marketing research and applications

    Jingmin Huang, Bowei Chen, Lan Luo, Shigang Yue, and Iadh Ounis. Dvm-car: A large-scale automotive dataset for visual marketing research and applications. In2022 IEEE In- ternational Conference on Big Data (Big Data), pages 4140–

  4. [12]

    Codenerf: Disentan- gled neural radiance fields for object categories

    Wonbong Jang and Lourdes Agapito. Codenerf: Disentan- gled neural radiance fields for object categories. In ICCV,

  5. [13]

    Real3d: Scaling up large reconstruction models with real- world images

    Hanwen Jiang, Qixing Huang, and Georgios Pavlakos. Real3d: Scaling up large reconstruction models with real- world images. arXiv preprint arXiv:2406.08479, 2024. 2

    2024 arXiv

  6. [14]

    Eschernet: A generative model for scalable view synthesis

    Xin Kong, Shikun Liu, Xiaoyang Lyu, Marwan Taher, Xi- aojuan Qi, and Andrew J Davison. Eschernet: A generative model for scalable view synthesis. CVPR, 2024. 2

    2024

  7. [15]

    Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model

    Jiahao Li, Hao Tan, Kai Zhang, Zexiang Xu, Fujun Luan, Yinghao Xu, Yicong Hong, Kalyan Sunkavalli, Greg Shakhnarovich, and Sai Bi. Instant3d: Fast text-to-3d with sparse-view generation and large reconstruction model. ICLR, 2024. 2

    2024

  8. [16]

    One-2-3-45++: Fast single image to 3d objects with consistent multi-view generation and 3d diffusion

    Minghua Liu, Ruoxi Shi, Linghao Chen, Zhuoyang Zhang, Chao Xu, Xinyue Wei, Hansheng Chen, Chong Zeng, Ji- ayuan Gu, and Hao Su. One-2-3-45++: Fast single image to 3d objects with consistent multi-view generation and 3d diffusion. arXiv preprint arXiv:2311.07885, 2023. 2

    2023 arXiv

  9. [17]

    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. NeurIPS, 2023. 2, 4

    2023

  10. [18]

    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 ICCV, 2023. 2, 3, 6

    2023

  11. [19]

    Car-studio: Learning car radiance fields from single- view and endless in-the-wild images

    Tianyu Liu, Hao Zhao, Yang Yu, Guyue Zhou, and Ming Liu. Car-studio: Learning car radiance fields from single- view and endless in-the-wild images. arXiv preprint arXiv:2307.14009, 2023. 3

    2023 arXiv

  12. [20]

    Syncdreamer: Gen- erating multiview-consistent images from a single-view im- age

    Yuan Liu, Cheng Lin, Zijiao Zeng, Xiaoxiao Long, Lingjie Liu, Taku Komura, and Wenping Wang. Syncdreamer: Gen- erating multiview-consistent images from a single-view im- age. arXiv preprint arXiv:2309.03453, 2023. 2

    2023 arXiv

  13. [21]

    Wonder3d: Sin- gle image to 3d using cross-domain diffusion.arXiv preprint arXiv:2310.15008, 2023

    Xiaoxiao Long, Yuan-Chen Guo, Cheng Lin, Yuan Liu, Zhiyang Dou, Lingjie Liu, Yuexin Ma, Song-Hai Zhang, Marc Habermann, Christian Theobalt, et al. Wonder3d: Sin- gle image to 3d using cross-domain diffusion.arXiv preprint arXiv:2310.15008, 2023. 2

    arXiv

  14. [22]

    Computing rectifying homographies for stereo vision

    Charles Loop and Zhengyou Zhang. Computing rectifying homographies for stereo vision. In Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pat- tern Recognition (Cat. No PR00149), pages 125–131. IEEE,

    1999

  15. [23]

    Im-3d: Iterative multiview diffusion and re- construction for high-quality 3d generation

    Luke Melas-Kyriazi, Iro Laina, Christian Rupprecht, Na- talia Neverova, Andrea Vedaldi, Oran Gafni, and Filippos Kokkinos. Im-3d: Iterative multiview diffusion and re- construction for high-quality 3d generation. arXiv preprint arXiv:2402.08682, 2024. 2

    2024 arXiv

  16. [24]

    Autorf: Learning 3d object radiance fields from single view observations

    Norman M ¨uller, Andrea Simonelli, Lorenzo Porzi, Samuel Rota Bul `o, Matthias Nießner, and Peter Kontschieder. Autorf: Learning 3d object radiance fields from single view observations. In CVPR, 2022. 2, 3, 4, 5, 6, 11

    2022

  17. [25]

    Pivotal tuning for latent-based editing of real im- ages

    Daniel Roich, Ron Mokady, Amit H Bermano, and Daniel Cohen-Or. Pivotal tuning for latent-based editing of real im- ages. ACM Transactions on graphics (TOG) , 42(1):1–13,

  18. [26]

    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, 2022. 2

    2022

  19. [27]

    Gina-3d: Learning to generate implicit neural as- sets in the wild

    Bokui Shen, Xinchen Yan, Charles R Qi, Mahyar Najibi, Boyang Deng, Leonidas Guibas, Yin Zhou, and Dragomir Anguelov. Gina-3d: Learning to generate implicit neural as- sets in the wild. In CVPR, 2023. 2, 3, 5

    2023

  20. [28]

    Zero123++: a single image to consistent multi-view dif- fusion base model

    Ruoxi Shi, Hansheng Chen, Zhuoyang Zhang, Minghua Liu, Chao Xu, Xinyue Wei, Linghao Chen, Chong Zeng, and Hao Su. Zero123++: a single image to consistent multi-view dif- fusion base model. arXiv preprint arXiv:2310.15110, 2023. 2, 4

    arXiv

  21. [29]

    Mvdream: Multi-view diffusion for 3d gen- eration

    Yichun Shi, Peng Wang, Jianglong Ye, Mai Long, Kejie Li, and Xiao Yang. Mvdream: Multi-view diffusion for 3d gen- eration. arXiv preprint arXiv:2308.16512, 2023. 2, 8

    2023 arXiv

  22. [30]

    Scalability in perception for autonomous driving: Waymo open dataset

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Yuning Chai, Benjamin Caine, et al. Scalability in perception for autonomous driving: Waymo open dataset. In CVPR,

  23. [31]

    Lgm: Large multi-view gaussian model for high-resolution 3d content creation

    Jiaxiang Tang, Zhaoxi Chen, Xiaokang Chen, Tengfei Wang, Gang Zeng, and Ziwei Liu. Lgm: Large multi-view gaussian model for high-resolution 3d content creation. ECCV, 2024. 2, 6

    2024

  24. [32]

    Mvdiffusion++: A dense high- resolution multi-view diffusion model for single or sparse- view 3d object reconstruction

    Shitao Tang, Jiacheng Chen, Dilin Wang, Chengzhou Tang, Fuyang Zhang, Yuchen Fan, Vikas Chandra, Yasutaka Fu- rukawa, and Rakesh Ranjan. Mvdiffusion++: A dense high- resolution multi-view diffusion model for single or sparse- view 3d object reconstruction. ECCV, 2024. 2

    2024

  25. [33]

    Neurad: Neural rendering for autonomous driving

    Adam Tonderski, Carl Lindstr ¨om, Georg Hess, William Ljungbergh, Lennart Svensson, and Christoffer Petersson. Neurad: Neural rendering for autonomous driving. CVPR,

  26. [34]

    Physically-based lighting for 3d generative models of cars

    Nicol ´as Violante, Alban Gauthier, Stavros Diolatzis, Thomas Leimk¨uhler, and George Drettakis. Physically-based lighting for 3d generative models of cars. In Computer Graphics Fo- rum, page e15011. Wiley Online Library, 2024. 3

    2024

  27. [35]

    Sv3d: Novel multi-view syn- thesis and 3d generation from a single image using latent video diffusion

    Vikram V oleti, Chun-Han Yao, Mark Boss, Adam Letts, David Pankratz, Dmitry Tochilkin, Christian Laforte, Robin Rombach, and Varun Jampani. Sv3d: Novel multi-view syn- thesis and 3d generation from a single image using latent video diffusion. In ECCV, 2024. 2

    2024

  28. [36]

    Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction

    Peng Wang, Hao Tan, Sai Bi, Yinghao Xu, Fujun Luan, Kalyan Sunkavalli, Wenping Wang, Zexiang Xu, and Kai Zhang. Pf-lrm: Pose-free large reconstruction model for joint pose and shape prediction. arXiv preprint arXiv:2311.12024, 2023. 2

    2023 arXiv

  29. [37]

    Crm: Single image to 3d textured mesh with convolu- tional reconstruction model

    Zhengyi Wang, Yikai Wang, Yifei Chen, Chendong Xiang, Shuo Chen, Dajiang Yu, Chongxuan Li, Hang Su, and Jun Zhu. Crm: Single image to 3d textured mesh with convolu- tional reconstruction model. In ECCV, 2024. 2

    2024

  30. [38]

    Novel view synthesis with diffusion models

    Daniel Watson, William Chan, Ricardo Martin-Brualla, Jonathan Ho, Andrea Tagliasacchi, and Mohammad Norouzi. Novel view synthesis with diffusion models. arXiv preprint arXiv:2210.04628, 2022. 2

    2022 arXiv

  31. [39]

    Consistent123: Improve consistency for one image to 3d object synthesis

    Haohan Weng, Tianyu Yang, Jianan Wang, Yu Li, Tong Zhang, CL Chen, and Lei Zhang. Consistent123: Improve consistency for one image to 3d object synthesis. arXiv preprint arXiv:2310.08092, 2023. 2

    2023 arXiv

  32. [40]

    Harmonyview: Harmonizing consis- tency and diversity in one-image-to-3d

    Sangmin Woo, Byeongjun Park, Hyojun Go, Jin-Young Kim, and Changick Kim. Harmonyview: Harmonizing consis- tency and diversity in one-image-to-3d. In CVPR, 2024. 2

    2024

  33. [41]

    Mars: An instance-aware, mod- ular and realistic simulator for autonomous driving

    Zirui Wu, Tianyu Liu, Liyi Luo, Zhide Zhong, Jianteng Chen, Hongmin Xiao, Chao Hou, Haozhe Lou, Yuantao Chen, Runyi Yang, et al. Mars: An instance-aware, mod- ular and realistic simulator for autonomous driving. In CAAI International Conference on Artificial Intelligence, pages 3–

  34. [42]

    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. 3

    2024 arXiv

  35. [43]

    Discoscene: Spatially disentangled generative radiance fields for controllable 3d- aware scene synthesis

    Yinghao Xu, Menglei Chai, Zifan Shi, Sida Peng, Ivan Sko- rokhodov, Aliaksandr Siarohin, Ceyuan Yang, Yujun Shen, Hsin-Ying Lee, Bolei Zhou, et al. Discoscene: Spatially disentangled generative radiance fields for controllable 3d- aware scene synthesis. In CVPR, 2023. 2, 3, 5, 6, 11

    2023

  36. [44]

    Dmv3d: Denoising multi-view diffu- sion using 3d large reconstruction model

    Yinghao Xu, Hao Tan, Fujun Luan, Sai Bi, Peng Wang, Ji- ahao Li, Zifan Shi, Kalyan Sunkavalli, Gordon Wetzstein, Zexiang Xu, et al. Dmv3d: Denoising multi-view diffu- sion using 3d large reconstruction model. arXiv preprint arXiv:2311.09217, 2023. 2

    2023 arXiv

  37. [45]

    Grm: Large gaussian reconstruction model for ef- ficient 3d reconstruction and generation

    Yinghao Xu, Zifan Shi, Wang Yifan, Hansheng Chen, Ceyuan Yang, Sida Peng, Yujun Shen, and Gordon Wet- zstein. Grm: Large gaussian reconstruction model for ef- ficient 3d reconstruction and generation. ECCV, 2024. 2

    2024

  38. [46]

    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 CVPR, 2023. 1, 8, 12

    2023

  39. [47]

    pixelNeRF: Neural radiance fields from one or few images

    Alex Yu, Vickie Ye, Matthew Tancik, and Angjoo Kanazawa. pixelNeRF: Neural radiance fields from one or few images. In CVPR, 2021. 3

    2021

  40. [48]

    Gs-lrm: Large recon- struction model for 3d gaussian splatting

    Kai Zhang, Sai Bi, Hao Tan, Yuanbo Xiangli, Nanxuan Zhao, Kalyan Sunkavalli, and Zexiang Xu. Gs-lrm: Large recon- struction model for 3d gaussian splatting. In ECCV, 2024. 2

    2024

  41. [49]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 586–595, 2018. 6

    2018

  42. [50]

    Free3d: Consis- tent novel view synthesis without 3d representation

    Chuanxia Zheng and Andrea Vedaldi. Free3d: Consis- tent novel view synthesis without 3d representation. arXiv preprint arXiv:2312.04551, 2023. 1, 2, 3, 5, 6, 11

    2023 arXiv

  43. [51]

    Triplane meets gaussian splatting: Fast and generalizable single-view 3d reconstruction with transformers

    Zi-Xin Zou, Zhipeng Yu, Yuan-Chen Guo, Yangguang Li, Ding Liang, Yan-Pei Cao, and Song-Hai Zhang. Triplane meets gaussian splatting: Fast and generalizable single-view 3d reconstruction with transformers. In CVPR, 2024. 2 Supplementary Material x y x yz z x y x yz z ObjaverseR...

    2024

  44. [52]

    absolute pose

    Supportive Explanations More explanations on relative pose vs. absolute pose. The reason that this matter arises lies in that, the azimuth angles of the Objaverse data are not defined w.r.t a common reference, as illustrated in Fig. 10. Specifically, the origin of the object c...

  45. [53]

    We provide addi- tional qualitative comparisons with AutoRF, DisCoScene, and Free3D in Fig

    Additional Results Additional qualitative comparisons. We provide addi- tional qualitative comparisons with AutoRF, DisCoScene, and Free3D in Fig. 12, with three examples to demon- strate the effectiveness of our method. Example (a) shows that all methods can handle a simple p...

Pith tools

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