Few-Shot Generalization for Single-Image 3D Reconstruction via Priors

Bharath Hariharan; Bram Wallace

REVIEW 2 major objections 6 minor 37 references

Few-Shot Generalization for Single-Image 3D Reconstruction via Priors

T0 review · 2 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read This paper argues that single-image 3D reconstruction can generalize to a never-seen object class using only one 3D shape from that class, supplied as an averaged prior at test time.

desk verdict Useful few-shot 3D reconstruction idea with honest ablations, but the headline 'one shape' claim is undercut by prior-shape variance. read the letter →

arxiv 1909.01205 v1 pith:TDSLGSYK submitted 2019-09-03 cs.CV cs.LGstat.ML

Bram Wallace , Bharath Hariharan This is my paper · ORCID

classification cs.CVcs.LGstat.ML

keywords single-view3Dreconstructionfew-shotlearningcategorypriorsvoxelgridsiterativerefinementtransfershapemulti-view

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

This paper tries to show that single-image 3D reconstruction can be extended to object classes never seen in training, using only one to a few 3D shapes from the new class and no images of it. Instead of learning a direct image-to-shape map, the network learns to refine a category-specific prior shape supplied at test time. The prior is just the average of a few voxelized shapes from the new category. On held-out categories the approach improves average intersection-over-union over a category-agnostic baseline (about 0.38–0.40 versus 0.36) and stays competitive with baselines that get the advantage of finetuning on images of the new class. A reader should care because it suggests category knowledge can be cleanly separated from image understanding and swapped in at runtime.

What carries the argument

The load-bearing object is the category prior as input: a floating-point voxel grid computed as the mean of k voxelized shapes from the target class. It carries category-specific structure (for example, the elongated shape of a rifle or vessel) while the network supplies category-agnostic refinement from the image. The architecture is a dual-encoder network: a 2D convolutional image encoder and a 3D convolutional shape encoder produce 128-dimensional embeddings that are added, and a decoder generates the refined voxel grid. Because prior and output share the same representation, the model can iterate, feeding its own prediction back as the prior. The training procedure mirrors this: during training the network is fed averaged same-category priors, and for iterative variants the output of one forward pass becomes the input of the next.

What would settle it

Present the trained 1-shot model with a novel-class image and the correct 1-shot prior, but replace the image with a uniform gray image of the same size; if the output IoU stays near the prior-only guess of about 0.23 instead of the reported 0.38, the image branch is not driving the reconstruction and the claimed image-driven refinement would not be supported.

Watch

Extended reading notes

Core claim

The central claim is that a single image can be converted into a 3D voxel reconstruction of an object from a class the network has never seen, provided the network is handed a category prior. The prior is a voxel grid computed by averaging the available 3D shapes from that class, and it can be as small as one shape. During training on base classes, the network sees pairs of an image and a same-category averaged prior, and learns a category-agnostic refinement function; at test time the prior is swapped for the novel category's average. The refinement is done by two encoders, one for the image and one for the prior, whose 128-dimensional outputs are summed and decoded. Because the output has the same voxel format as the input prior, it can be fed back in and refined again, which the paper exploits for multi-view reconstruction without any multi-view training. The paper reports average IoU gains on novel classes over an image-only baseline and shows the gain is not simply regurgitating the prior.

Load-bearing premise

The system only delivers its improvement if, at test time, the object's category is known and a matching prior is supplied; when the prior comes from the wrong category, performance drops below the no-prior baseline.

Editorial extensions

If this is right

A novel object class can be added to a trained reconstruction system with negligible runtime cost: just load and average a handful of shapes; no retraining or category-specific weights are needed.
The same network, without any multi-view training, improves its reconstruction as more views are fed in by iterating the refinement, gaining close to 7 IoU points on novel classes over its single-view performance.
The gain on novel classes comes mostly from eliminating very poor reconstructions rather than improving already-good ones; on rifles, for example, the fraction of instances with IoU below 0.1 drops from over half to under 17 percent.
If the wrong category prior is supplied, performance falls below the category-agnostic baseline, so categorical knowledge at test time is required for the improvement.

Reading between the lines

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

The same prior-as-input design could transfer to other structured prediction problems, such as semantic segmentation or shape completion, wherever a cheap class-specific template exists and can be averaged.
The finding that wrong priors hurt more than no prior suggests the network may be amplifying prior mistakes rather than merely smoothing them; a testable extension would be to train with deliberately corrupted priors to make the refinement more robust.
If the prior is just a mean shape, one could explore richer priors, such as a small set of modes or a learned low-dimensional shape space, which might extend the approach to categories with high intra-class variation like lamps, where the paper finds smaller gains.

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.

Desk Editor's Note

read the letter

The core idea here is genuinely new: instead of training a direct image-to-shape mapping, train a category-agnostic refinement network that takes a category-averaged voxel prior as a second input. That lets you swap in a prior for a novel class at runtime with no retraining. The multi-view byproduct is a nice bonus. The paper is clearly written, the ablations are honest, and the wrong-prior experiments plus the naive-prior baseline give real evidence that the model is doing inference, not regurgitating the prior.

The main results hold up in the medium-shot regime. At 5-shot and above, the improvement over the category-agnostic baseline (0.40 vs 0.36) is consistent and the per-category analysis shows the gains are mostly real, though concentrated in vessels and rifles. The failure on cabinets and benches is sensible and discussed.

The soft spot is the 1-shot claim, and it is load-bearing. The paper says "as few as one 3D shape" and highlights 1-shot in the abstract, but the reported gain is 0.02 IoU (0.38 vs 0.36). Section 5.4.2 says that using different single prior shapes on the same image-target pair gives sigma ~ 0.05. That means the prior-shape draw is the dominant source of randomness, and the reported mean could easily flip sign. The reported sigma < 0.01 across runs covers only the training/data randomness, not the prior selection. This is a real gap and the stress-test note is correct.

Also missing: Zhang et al. (NeurIPS 2018) is the closest prior work on reconstructing unseen classes, and it is not included in the comparison tables. The finetuning baselines are trained with a fixed 200 iterations, which may disadvantage them, and the best iteration is selected post hoc. No code is released. These are fixable but they matter.

The phrase "significantly outperforms" is too strong for a 2-point gain with that variance. For 5-shot it is fairer, but the abstract's one-shot emphasis is not supported.

Overall: this is a solid, honest paper with a useful idea and a credible medium-shot result. The 1-shot claim needs better uncertainty quantification and the missing baseline should be addressed. It deserves peer review, not desk rejection, but it needs revision before acceptance.

Referee Report

2 major / 6 minor

Summary. The paper proposes a dual-encoder architecture for single-view 3D reconstruction in which a category-agnostic network refines a category-specific voxel prior provided at test time. The prior is the average of k 3D shapes from the target category; for novel categories no images and no retraining are used. Training is performed on seven ShapeNet base classes, and evaluation on six held-out categories shows k-shot gains over a category-agnostic image-only baseline and competitiveness with finetuning baselines. The paper also shows that feeding the output back as a new prior enables multi-view reconstruction without multi-view training, and it includes ablations with naive priors, incorrect priors, and PASCAL 3D+ fine-tuning.

Significance. If the results hold, the paper makes a useful empirical contribution: a simple, training-free mechanism for adapting single-view 3D reconstruction to novel classes using only a handful of 3D examples, with careful controls showing that the prior, not the learned model, carries the category-specific information. The wrong-prior experiment and the naive-prior baseline are exactly the right controls, and the per-category analysis is informative. The main weakness is that the strongest headline claim—improvement from a single 3D shape—is not supported with uncertainty that accounts for randomness in selecting the prior shape; the reported within-run sigma is an order of magnitude smaller than the prior-draw sigma reported in the same paper. The finetuning baseline protocol also needs justification.

major comments (2)

[Sec. 5.2, Table 1; Sec. 5.4.2] The headline one-shot claim lacks appropriate uncertainty quantification. Table 1 reports 0.38 IoU for a 1-shot prior versus 0.36 for the image-only baseline and states sigma_IoU < 0.01 across 3-5 runs, but Section 5.4.2 reports that using different 1-shot shapes on the same image-target pair yields sigma approximately 0.05. In the 1-shot regime the prior is a single randomly selected novel-class shape, so the reported mean depends on which shape was drawn, and the 0.02 gain is roughly 2.5 times smaller than the prior-draw sigma. Table 5 further shows that the gain is not uniform (benches and cabinets show 0.0% improvement, lamps 5.6%). Please report a confidence interval or significance test that treats the prior-shape draw as a random effect, specify how the k shapes are sampled, and either establish that the one-shape gain is robust or soften the abstract's one-shape claim.
[Sec. 5.1, Baselines; Table 1] The finetuning baselines are trained with a fixed budget of 200 SGD iterations at learning rate 0.005 without early stopping or validation-based selection. The abstract claims that the method 'remains competitive with more sophisticated baselines that finetune,' but an undertrained baseline could make this claim appear stronger than it is. Please justify the training budget (e.g., show performance versus iteration count, use early stopping on a validation split, or tune the learning rate), and report the batch size and number of examples seen so that the reader can judge whether the baselines have converged.

minor comments (6)

[Sec. 5.4.2, Table 5] The statement that the 1-shot naive guess 'performs significantly worse than both of our models' is contradicted by the Rifles row, where the naive guess (0.27) exceeds both the 3-Iteration (0.16) and 1-Iteration 1-Shot (0.19) models; please qualify this claim.
[Sec. 5.2, Table 5] The text says the model 'significantly outperforms the category-agnostic baseline across the board,' but Table 5 shows no improvement on benches and cabinets for the 1-iteration 1-shot model and a decrease on cabinets for the 3-iteration full-prior model; 'on average' would be accurate.
[Table 2] Table 2 is difficult to parse: pairs such as '0.34/0.37/0.380.38/0.38' are missing separators, and the training-prior/test-prior configuration for each column should be stated explicitly.
[Sec. 5.1; Sec. 5.4.3] Please state explicitly that the prior shapes for novel classes are drawn from the training split of those categories and are disjoint from the test examples, and clarify whether the '13 ShapeNet categories' in the PASCAL experiment refers to pretraining or finetuning.
[Throughout] There are numerous typos (e.g., 'taarget', 'thaan', 'neccessary', 'peformed') and inconsistent use of the ligature 'riﬂes'; a careful proofread is needed.
[Sec. 5.3, Table 4] The multi-view benefit is shown for the 3-iteration full-prior model, which is not the best single-view model (the 1-iteration 1-shot model is 0.39 and flat with more views); the paper should address this trade-off explicitly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: prior is an external held-out input, and the naive-prior/wrong-prior controls separate the prior's contribution from the learned refinement.

full rationale

Walking the derivation chain, the central claim (a category-agnostic network can refine a category-specific prior obtained from 1-25 held-out novel-class shapes, with no novel-class images and no retraining) is not equivalent to its inputs by construction. The prior is an external input, not a fitted parameter; the model is trained only on base classes, and novel-class priors are computed from held-out shapes without using test images or ground-truth shapes. The paper explicitly controls for the concern that the model might merely regurgitate the prior: the '1-Shot Guess' column in Table 5 shows the prior alone achieves mean IoU 0.23 vs 0.38-0.39 for the model, and Section 5.4.2 reports that feeding priors from wrong categories drops performance below the baseline, confirming the prior and the learned refinement are separable. The only self-citation is to Yang et al. [33] for the encoder/decoder design and base-class split, and that citation is not load-bearing for the few-shot claim. The statistical concern that the 1-shot gain (0.02) is comparable to the prior-draw sigma (~0.05) is a robustness/significance issue, not a circularity, and does not affect this verdict.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new theoretical entities. The central claim rests on two empirical assumptions: the availability of a matching category prior at test time, and the sufficiency of averaging a few shapes to build that prior. Free parameters are standard network and evaluation hyperparameters, plus a mild post hoc selection of iteration count.

free parameters (4)

IoU binarization threshold = 0.4
Used to convert continuous voxel probabilities to binary shapes before computing IoU. The paper states relative performances were maintained at other thresholds, so it is not load-bearing.
Number of refinement iterations for headline results = 1 for few-shot, 3 for multi-view
The best-performing iteration per model is underlined in Table 2 and the best model variant is used in Table 1; this is a post hoc selection on test data, a mild free choice.
Shape-encoder LeakyReLU slope = 0.3 (image encoder 0.01)
Hand-chosen activation slopes; typical hyperparameters not central to the claim.
Latent fusion dimension = 128
Dimension of the summed image and prior embeddings; standard architectural choice inherited from R2N2 and Yang et al.

assumptions (3)

domain assumption A correct category-specific prior is available at test time.
Load-bearing: the method's improvement disappears when a random-category prior is used (Fig. 8). The paper acknowledges this and argues category knowledge is a common assumption in prior single-view reconstruction work.
domain assumption Averaging k voxelized shapes from a category yields a useful summary of that category's shape distribution.
The prior is constructed by this average and is never learned or optimized; the paper validates it empirically for k=1 to 25, but provides no formal argument that the mean voxel grid is a sufficient category statistic.
domain assumption ShapeNet's rendered images and 32x32x32 voxel grids are a valid training and evaluation substrate for the claimed generalization.
All conclusions are drawn on this benchmark; Sec. 5.4.3 notes real-world PASCAL data has very low shape variation and should be interpreted cautiously.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Few-Shot Generalization for Single-Image 3D Reconstruction via Priors." pith.science (2026). https://pith.science/paper/TDSLGSYK

@misc{pith2026190901205,
  author       = {Pith},
  title        = {Pith review of: Few-Shot Generalization for Single-Image 3D Reconstruction via Priors},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TDSLGSYK}},
  note         = {Machine review of arXiv:1909.01205}
}

read the original abstract

Recent work on single-view 3D reconstruction shows impressive results, but has been restricted to a few fixed categories where extensive training data is available. The problem of generalizing these models to new classes with limited training data is largely open. To address this problem, we present a new model architecture that reframes single-view 3D reconstruction as learnt, category agnostic refinement of a provided, category-specific prior. The provided prior shape for a novel class can be obtained from as few as one 3D shape from this class. Our model can start reconstructing objects from the novel class using this prior without seeing any training image for this class and without any retraining. Our model outperforms category-agnostic baselines and remains competitive with more sophisticated baselines that finetune on the novel categories. Additionally, our network is capable of improving the reconstruction given multiple views despite not being trained on task of multi-view reconstruction.

Figures

Figures reproduced from arXiv: 1909.01205 by the authors.

**Figure 2.** Our model is dual-input. The first input is an image encoded using the exact same architecture as 3D-R2N2 [ [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** The averaged shapes of the entire training dataset for [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 5.** Figure 5: Examples of ground-truth and predicted shapes from an [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]

**Figure 4.** Figure 4: Performance of the 1-iteration 1-shot-trained model [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

**Figure 6.** Figure 6: Here we plot the IoUs in increasing order for each model [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

**Figure 7.** Figure 7: Scatter plots of model performances vs each other for [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]

**Figure 8.** Figure 8: Performance across iterations with the model being fed [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 35 canonical work pages

[1]

Mart ´ın Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S. Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Ian Goodfellow, Andrew Harp, Geoffrey Irving, Michael Is- ard, Yangqing Jia, Rafal Jozefowicz, Lukasz Kaiser, Man- junath Kudlur, Josh Levenberg, Dan Man ´e, Rajat Monga, Sherry Moore, Derek Mur...

work page 2015
[2]

Human pose estimation with iterative error feedback

Joao Carreira, Pulkit Agrawal, Katerina Fragkiadaki, and Ji- tendra Malik. Human pose estimation with iterative error feedback. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016. 2, 3

work page 2016
[3]

Learning priors for se- mantic 3d reconstruction

Ian Cherabier, Johannes L Schonberger, Martin R Oswald, Marc Pollefeys, and Andreas Geiger. Learning priors for se- mantic 3d reconstruction. In Proceedings of the European Conference on Computer Vision (ECCV) , pages 314–330,

work page
[4]

Keras.https://keras.io, 2015

Franc ¸ois Chollet et al. Keras.https://keras.io, 2015. 3

work page 2015
[5]

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

Christopher B Choy, Danfei Xu, JunYoung Gwak, Kevin Chen, and Silvio Savarese. 3d-r2n2: A uniﬁed approach for single and multi-view 3d object reconstruction. In European conference on computer vision , pages 628–644. Springer,

work page
[6]

Accurate, dense, and ro- bust multiview stereopsis

Yasutaka Furukawa and Jean Ponce. Accurate, dense, and ro- bust multiview stereopsis. TPAMI, 32(8):1362–1376, 2010. 2

work page 2010
[7]

Kim, Bryan Russell, and Mathieu Aubry

Thibault Groueix, Matthew Fisher, Vladimir G. Kim, Bryan Russell, and Mathieu Aubry. AtlasNet: A Papier-Mˆach´e Ap- proach to Learning 3D Surface Generation. In CVPR, 2018. 2, 8

work page 2018
[8]

Low-shot visual recognition by shrinking and hallucinating features

Bharath Hariharan and Ross Girshick. Low-shot visual recognition by shrinking and hallucinating features. In The IEEE International Conference on Computer Vision (ICCV), Oct 2017. 3

work page 2017

Show all 37 references

[9]

Efros, and Jitendra Malik

Angjoo Kanazawa, Shubham Tulsiani, Alexei A. Efros, and Jitendra Malik. Learning category-speciﬁc mesh reconstruc- tion from image collections. In ECCV, 2018. 2

2018
[10]

Learning a multi-view stereo machine

Abhishek Kar, Christian H ¨ane, and Jitendra Malik. Learning a multi-view stereo machine. In Advances in neural infor- mation processing systems, pages 365–376, 2017. 2

2017
[11]

Learning a multi-view stereo machine

Abhishek Kar, Christian H ¨ane, and Jitendra Malik. Learning a multi-view stereo machine. 2017. 2, 6

2017
[12]

Learning view priors for single-view 3d reconstruction

Hiroharu Kato and Tatsuya Harada. Learning view priors for single-view 3d reconstruction. CoRR, abs/1811.10719,

arXiv
[13]

A theory of shape by space carving

Kiriakos N Kutulakos and Steven M Seitz. A theory of shape by space carving. International journal of computer vision , 38(3):199–218, 2000. 2

2000
[14]

Learning from one example through shared densities on transforms

Erik G Miller, Nicholas E Matsakis, and Paul A Viola. Learning from one example through shared densities on transforms. In CVPR, 2000. 3

2000
[15]

An- drew Bagnell

Stephane Ross, Daniel Munoz, Martial Hebert, and J. An- drew Bagnell. Learning message-passing inference ma- chines for structured prediction. In CVPR, 2011. 2, 3

2011
[16]

Prototyp- ical networks for few-shot learning

Jake Snell, Kevin Swersky, and Richard S Zemel. Prototyp- ical networks for few-shot learning. In NIPS, 2017. 2

2017
[17]

Octree generating networks: Efﬁcient convolutional archi- tectures for high-resolution 3d outputs

Maxim Tatarchenko, Alexey Dosovitskiy, and Thomas Brox. Octree generating networks: Efﬁcient convolutional archi- tectures for high-resolution 3d outputs. InProceedings of the IEEE International Conference on Computer Vision , pages 2088–2096, 2017. 2

2017
[18]

What do single-view 3d reconstruction networks learn? In Proceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 3405–3414, 2019

Maxim Tatarchenko, Stephan R Richter, Ren ´e Ranftl, Zhuwen Li, Vladlen Koltun, and Thomas Brox. What do single-view 3d reconstruction networks learn? In Proceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 3405–3414, 2019. 2, 8

2019
[19]

Auto-context and its application to high-level vision tasks

Zhuowen Tu. Auto-context and its application to high-level vision tasks. In CVPR, pages 1–8, 2008. 2, 3

2008
[20]

Efros, and Jitendra Malik

Shubham Tulsiani, Alexei A. Efros, and Jitendra Malik. Multi-view consistency as supervisory signal for learning shape and pose prediction. In CVPR, 2018. 2

2018
[21]

Matching networks for one shot learning

Oriol Vinyals, Charles Blundell, Timothy Lillicrap, Daan Wierstra, et al. Matching networks for one shot learning. In Advances in neural information processing systems , pages 3630–3638, 2016. 2

2016
[22]

3densinet: A ro- bust neural network architecture towards 3d volumetric ob- ject prediction from 2d image

Meng Wang, Lingjing Wang, and Yi Fang. 3densinet: A ro- bust neural network architecture towards 3d volumetric ob- ject prediction from 2d image. In Proceedings of the 25th ACM international conference on Multimedia , pages 961–
[23]

Pixel2mesh: Generating 3d mesh models from single rgb images

Nanyang Wang, Yinda Zhang, Zhuwen Li, Yanwei Fu, Wei Liu, and Yu-Gang Jiang. Pixel2mesh: Generating 3d mesh models from single rgb images. In Proceedings of the Euro- pean Conference on Computer Vision (ECCV), pages 52–67,
[24]

O-cnn: Octree-based convolutional neu- ral networks for 3d shape analysis

Peng-Shuai Wang, Yang Liu, Yu-Xiao Guo, Chun-Yu Sun, and Xin Tong. O-cnn: Octree-based convolutional neu- ral networks for 3d shape analysis. ACM Transactions on Graphics (TOG), 36(4):72, 2017. 2

2017
[25]

3dn: 3d deformation network

Weiyue Wang, Duygu Ceylan, Radomir Mech, and Ulrich Neumann. 3dn: 3d deformation network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1038–1046, 2019. 2

2019
[26]

Marrnet: 3d shape recon- struction via 2.5 d sketches

Jiajun Wu, Yifan Wang, Tianfan Xue, Xingyuan Sun, Bill Freeman, and Josh Tenenbaum. Marrnet: 3d shape recon- struction via 2.5 d sketches. In Advances in neural informa- tion processing systems, pages 540–550, 2017. 2

2017
[27]

Learning a probabilistic latent space of object shapes via 3d generative-adversarial modeling

Jiajun Wu, Chengkai Zhang, Tianfan Xue, Bill Freeman, and Josh Tenenbaum. Learning a probabilistic latent space of object shapes via 3d generative-adversarial modeling. InAd- vances in neural information processing systems, pages 82– 90, 2016. 2

2016
[28]

Freeman, and Joshua B

Jiajun Wu, Chengkai Zhang, Xiuming Zhang, Zhoutong Zhang, William T. Freeman, and Joshua B. Tenenbaum. Learning shape priors for single-view 3d completion and re- construction. In The European Conference on Computer Vi- sion (ECCV), September 2018. 2

2018
[29]

Beyond pascal: A benchmark for 3d object detection in the wild

Yu Xiang, Roozbeh Mottaghi, and Silvio Savarese. Beyond pascal: A benchmark for 3d object detection in the wild. In IEEE Winter Conference on Applications of Computer Vision (WACV), pages 75–82, 2014. 8

2014
[30]

Low-shot learning from imaginary data

Yu xiong Wang, Ross Girshick, Martial Herbert, and Bharath Hariharan. Low-shot learning from imaginary data. In CVPR, 2018. 2

2018
[31]

Empirical evaluation of rectiﬁed activations in convolutional network

Bing Xu, Naiyan Wang, Tianqi Chen, and Mu Li. Empirical evaluation of rectiﬁed activations in convolutional network. arXiv preprint arXiv:1505.00853, 2015. 3

2015 arXiv
[32]

Perspective transformer nets: Learning single- view 3d object reconstruction without 3d supervision

Xinchen Yan, Jimei Yang, Ersin Yumer, Yijie Guo, and Honglak Lee. Perspective transformer nets: Learning single- view 3d object reconstruction without 3d supervision. In NeurIPS, pages 1696–1704, 2016. 2

2016
[33]

Learning single-view 3d reconstruction with limited pose supervision

Guandao Yang, Yin Cui, Serge Belongie, and Bharath Hari- haran. Learning single-view 3d reconstruction with limited pose supervision. In Proceedings of the European Confer- ence on Computer Vision (ECCV) , pages 86–101, 2018. 1, 2, 3, 4, 8

2018
[34]

Fold- ingnet: Point cloud auto-encoder via deep grid deformation

Yaoqing Yang, Chen Feng, Yiru Shen, and Dong Tian. Fold- ingnet: Point cloud auto-encoder via deep grid deformation. In CVPR, 2018. 2

2018
[35]

Adadelta: an adaptive learning rate method

Matthew D Zeiler. Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701, 2012. 3

2012 arXiv
[36]

Learning to re- construct shapes from unseen classes

Xiuming Zhang, Zhoutong Zhang, Chengkai Zhang, Josh Tenenbaum, Bill Freeman, and Jiajun Wu. Learning to re- construct shapes from unseen classes. In S. Bengio, H. Wal- lach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett, editors, Advances in Neural Information Pro...
[37]

3d-prnn: Generating shape primitives with recurrent neural networks

Chuhang Zou, Ersin Yumer, Jimei Yang, Duygu Ceylan, and Derek Hoiem. 3d-prnn: Generating shape primitives with recurrent neural networks. In The IEEE International Con- ference on Computer Vision (ICCV), 2017. 2

2017

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

Mart ´ın Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S. Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Ian Goodfellow, Andrew Harp, Geoffrey Irving, Michael Is- ard, Yangqing Jia, Rafal Jozefowicz, Lukasz Kaiser, Man- junath Kudlur, Josh Levenberg, Dan Man ´e, Rajat Monga, Sherry Moore, Derek Mur...

work page 2015

[2] [2]

Human pose estimation with iterative error feedback

Joao Carreira, Pulkit Agrawal, Katerina Fragkiadaki, and Ji- tendra Malik. Human pose estimation with iterative error feedback. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016. 2, 3

work page 2016

[3] [3]

Learning priors for se- mantic 3d reconstruction

Ian Cherabier, Johannes L Schonberger, Martin R Oswald, Marc Pollefeys, and Andreas Geiger. Learning priors for se- mantic 3d reconstruction. In Proceedings of the European Conference on Computer Vision (ECCV) , pages 314–330,

work page

[4] [4]

Keras.https://keras.io, 2015

Franc ¸ois Chollet et al. Keras.https://keras.io, 2015. 3

work page 2015

[5] [5]

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

Christopher B Choy, Danfei Xu, JunYoung Gwak, Kevin Chen, and Silvio Savarese. 3d-r2n2: A uniﬁed approach for single and multi-view 3d object reconstruction. In European conference on computer vision , pages 628–644. Springer,

work page

[6] [6]

Accurate, dense, and ro- bust multiview stereopsis

Yasutaka Furukawa and Jean Ponce. Accurate, dense, and ro- bust multiview stereopsis. TPAMI, 32(8):1362–1376, 2010. 2

work page 2010

[7] [7]

Kim, Bryan Russell, and Mathieu Aubry

Thibault Groueix, Matthew Fisher, Vladimir G. Kim, Bryan Russell, and Mathieu Aubry. AtlasNet: A Papier-Mˆach´e Ap- proach to Learning 3D Surface Generation. In CVPR, 2018. 2, 8

work page 2018

[8] [8]

Low-shot visual recognition by shrinking and hallucinating features

Bharath Hariharan and Ross Girshick. Low-shot visual recognition by shrinking and hallucinating features. In The IEEE International Conference on Computer Vision (ICCV), Oct 2017. 3

work page 2017