Efficient Object-centric Representation Learning with Pre-trained Geometric Prior

Alan F. Smeaton; Graham Healy; Ph\'uc H. Le Khac

REVIEW 3 major objections 5 minor 23 references

Efficient Object-centric Representation Learning with Pre-trained Geometric Prior

T0 review · 3 major / 5 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 This paper claims that a pre-trained geometric vision model, used as the reconstruction target for slot attention, improves object-centric segmentation in complex synthetic video, outperforming a semantic target and approaching…

desk verdict A useful, counterintuitive empirical result on CroCo features for object discovery, but the central geometric-prior claim is confounded with decoder and pipeline changes—needs a matched-target ablation and error bars before accepting. read the letter →

arxiv 2412.12331 v1 pith:ETFWU7BQ submitted 2024-12-16 cs.CV cs.MM

Phúc H. Le Khac , Graham Healy , Alan F. Smeaton This is my paper · ORCID

classification cs.CVcs.MM

keywords object-centricrepresentationlearningslotattentiongeometricpriorself-supervisedpre-trainingvideoobjectsegmentationattentionaldecoderweaklysupervised

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 argues that the target a slot-attention video model is trained to reconstruct determines how well it discovers objects, and that geometric features are a better target than semantic features. Training an object-centric video model to predict CroCo's pre-trained geometric representations, rather than DINO's semantic representations, raises foreground segmentation ARI-FG from 0.686 to 0.788 on MOVi-C and from 0.651 to 0.766 on MOVi-E, close to the depth-supervised SA Vi++ (0.843 and 0.823). This is achieved without explicit depth input, using a newly designed attentional slot decoder that performs a single decoding pass and reduces training memory from about 24 GB to about 6 GB. The practical stake is that object-centric learning, which has struggled to scale to complex scenes, might be bootstrapped from readily available geometric self-supervised models rather than requiring dense supervision.

What carries the argument

The load-bearing component is the Attentional Slot Decoder. Instead of a spatial broadcast decoder that reconstructs each slot independently, this decoder initialises a spatial grid of Fourier positional embeddings, adds a pooled global scene embedding, and runs one cross-attention operation in which the slot latents serve as keys and values. The softmax-normalised attention scores over the slots are directly used as alpha masks for segmentation, and a residual connection from positional embeddings lets the decoder place objects at the correct locations. Around this, the method uses Slot Attention as the slot encoder, CroCo's pre-trained ViT as both visual backbone and reconstruction target, and a mean-squared-error loss in the feature space. The efficiency gain comes from replacing the per-slot decoding passes of mixture-of-components decoders with a single set-based cross-attention pass.

What would settle it

Train the paper's exact pipeline but replace the CroCo feature target with DINO features (or vice versa, train DINOSAUR's pipeline with CroCo features) while keeping every other component identical. If ARI-FG does not move with the target, the geometric-prior claim is falsified; if it does, the claim is supported.

Watch

Extended reading notes

Core claim

The central claim is that the inherent geometric information in a pre-trained cross-view completion model (CroCo) provides a better training signal for object-centric discovery than the semantic information in DINO. Counterintuitively, even though DINO's attention maps visibly focus on foreground objects while CroCo's attention is diffuse, models trained to reconstruct CroCo features segment objects more accurately. The paper attributes this to geometry encoding 3D layout, occlusion, and motion cues that semantics lack. On MOVi-C and MOVi-E, the proposed method surpasses DINOSAUR (which uses DINO features) by roughly 10 absolute ARI-FG points and approaches SA Vi++, which uses explicit depth supervision. The authors also claim their attentional slot decoder reaches this performance with a single forward decoding pass and roughly a quarter of the training memory.

Load-bearing premise

The results hinge on the premise that the segmentation gains come from the geometric prediction target rather than from the new attentional decoder or the recurrent slot initialisation, since the comparison against DINOSAUR changes both the target and the architecture.

Editorial extensions

If this is right

If correct, object-centric video models can be trained to state-of-the-art segmentation without explicit depth, using only frozen pre-trained geometric features as the target.
The attentional slot decoder reduces training memory roughly fourfold relative to SA Vi's spatial broadcast decoder, which lowers the barrier to scaling slot attention to longer or higher-resolution videos.
The choice of reconstruction target—geometric versus semantic—can matter more than the objectness apparent in the pre-trained model's attention maps.
The method is modality-flexible: the same decoder works with RGB, optical flow, and pre-trained feature targets, suggesting a single architecture can suit various reconstruction signals.

Reading between the lines

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

A direct test that would isolate the geometric prior's contribution is to train the same attentional slot decoder with DINO and with CroCo targets under identical slot initialisation and training schedule; the paper's comparison does not fully control for decoder architecture, so part of the +10% may come from the decoder rather than the geometric prior.
If geometry is the key, then other geometric or 3D-aware pre-training objectives—for example depth-prediction or multi-view consistency—should produce comparable or better object discovery than CroCo, offering a cheap route to scale object-centric learning to real-world video.
The inflated object masks the paper observes (shadows included) hint that geometric features encode object extent rather than tight boundaries; fine-tuning the target with contour or edge information might sharpen segmentation without losing the geometric advantage.

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.

Desk Editor's Note

read the letter

Quick take: this is a competent empirical paper giving the object-centric video community two useful things—a one-pass attentional slot decoder that cuts training memory roughly 4x (24GB to 6GB), and evidence that a reconstruction target with geometric signal (CroCo features) beats a semantic one (DINO/DINOSAUR) on MOVi-C/E object discovery. The decoder is a sensible set-based cross-attention variant with positional queries plus a broadcast global embedding; the memory reduction is real, and the ablations show the positional and global-embedding choices matter.

The result is more interesting because the paper shows CroCo's attention maps are diffuse and non-object-like, yet slot attention on CroCo features yields higher ARI-FG than DINO features. That counterintuitive finding is worth taking seriously.

The soft spot is the central comparison. Table II varies prediction target, decoder architecture, and temporal pipeline jointly. DINOSAUR is adapted from its image version with its own decoder, while the proposed method uses the attentional decoder plus recurrent slot initialization inherited from SA Vi. No ablation holds the decoder fixed and swaps only CroCo vs DINO as target. So the observed +10% ARI-FG over DINOSAUR could come from the decoder's cross-attention mechanism or the recurrent slot carry-over, not from geometric information. The paper does include useful ablations of its own decoder, but not this key target swap. There are also no error bars or multiple seeds; the margins are large, but seed variation is unquantified.

Minor points: SA Vi with RGB pixels is a weak baseline, so beating it is not surprising; the meaningful comparisons are DINOSAUR and SA Vi++. The paper is honest about failure cases and error accumulation. The citation pattern is appropriate, with no self-citation inflation.

Bottom line: the empirical result is probably in the right direction, but the headline claim is conditional. It deserves a serious referee: send it to review, and require the matched-decoder target ablation, multi-seed results, and ideally code. If the target swap confirms the geometric-prior advantage, it becomes a solid contribution.

For whom: researchers working on slot attention, object-centric video, and self-supervised feature targets. Recommendation: engage, with the ablation as the gate.

Referee Report

3 major / 5 minor

Summary. The paper proposes an object-centric video representation learning framework that uses a pre-trained CroCo encoder's geometric feature map as the reconstruction target for a Slot Attention module. The main architectural contribution is an Attentional Slot Decoder that uses a 2D Fourier-positional query grid, a global scene embedding, and cross-attention over slot latents, with the attention scores directly serving as segmentation masks. On MOVi-C and MOVi-E, the method reports ARI-FG values of 0.788 and 0.766, respectively, exceeding SA Vi and DINOSAUR and approaching the depth-supervised SA Vi++ (0.843/0.823), while requiring about 6 GB of training memory versus 24 GB for SA Vi. The paper also includes decoder ablations, qualitative results, and an analysis of attention maps from several pre-trained models.

Significance. If the results are robust, the paper would support a useful and somewhat counterintuitive claim: a geometric self-supervised representation (CroCo) can serve as a better reconstruction target for slot-attention object discovery than a semantic representation (DINO), and the proposed decoder gives a major memory saving. The paper has several concrete strengths: the decoder ablations in Table I (positional embedding, global scene embedding) isolate specific design choices, the evaluation uses external ground-truth masks and ARI-FG, so there is no training/evaluation circularity, and the paper honestly documents that CroCo attention maps do not show object grouping. However, the central causal claim about the geometric prior is currently supported only by a cross-method comparison that changes several variables at once, and the main quantitative comparisons are single-run point estimates. The potential value is real, but the evidence needs targeted additions before the claim is established.

major comments (3)

[Section IV-B, Table II] The central claim that the geometric CroCo target, rather than the decoder or training pipeline, causes the +10% ARI-FG gain over DINOSAUR is not identifiable from the reported experiments. The rows of Table II differ jointly in prediction target, decoder architecture, and temporal slot-initialisation: DINOSAUR is adapted from its image version with its own decoder, while the proposed method uses the Attentional Slot Decoder from Section III-B together with the recurrent slot initialisation inherited from SA Vi. No ablation holds the decoder and pipeline fixed and varies only the prediction target. A minimal test would train the proposed pipeline with DINO features as the target and compare it to the CroCo-target version (or, symmetrically, train DINOSAUR's decoder with CroCo features). Without such an ablation, the improvement in Table II could be due to the decoder's cross-attention, the residual positional injection, or the recurrent slot carry-over, rather than to geometric information. This is load-bearing because the paper's title and Section IV-B explicitly attribute the improvement to geometric versus semantic representations.
[Section IV-B, Table II] All reported values in Table II, including the headline margins (0.788 vs 0.686 on MOVi-C and 0.766 vs 0.651 on MOVi-E), are point estimates with no error bars, no number of seeds, and no significance test. Slot Attention training is known to be sensitive to seed variation, so the observed margins could shrink or change in relative ordering. The paper should report means and standard deviations over at least three seeds, or provide a clear statement of variance in the appendix. This is not merely a presentation issue because the claims 'we surpass' and '+10% improvement' are stated in absolute terms from single runs.
[Section III-A(c), Section IV-B] The paper's own analysis in Section III-A(c) shows that CroCo attention maps are diffuse and do not exhibit object-grouping behavior, unlike DINO and MSN. This observation is not a direct argument against the geometric-prior hypothesis, but it also provides no supporting evidence for the mechanism by which geometric features improve object discovery. The paper says the improvement is 'attributed to the distinctive characteristics of geometric and semantic representations,' yet no experiment measures or ablates the geometric content of the target features. The matched-target ablation proposed in the first comment would be the minimal way to support this mechanism; without it, the mechanism remains speculative.

minor comments (5)

[Table I] The row labels '- Pos.' and '- CLS' are not explained in the table caption. The text describes removing positional information and removing the global scene representation, but the table abbreviates the latter as 'CLS'; please define whether 'CLS' refers to the global embedding taken from the CLS token and confirm the correspondence with the text.
[Table II] The SA Vi++ values (0.843 and 0.823) are listed without a source citation in the table. Please state whether these numbers are taken directly from [17] or re-evaluated under the same protocol, and cite the specific source or appendix.
[Abstract and Section IV] The abstract calls the method 'weakly-supervised,' but the paper also describes the task as 'unsupervised discovery' and uses bounding boxes only for conditional slot initialisation. Please clarify the terminology and reconcile it with the conditional object-centric learning setting inherited from SA Vi.
[Section III-C] The loss is written as L = ||Y - V||_2^2 without stating whether the norm is averaged over spatial positions and feature channels. Please clarify the exact MSE normalization used in training.
[Conclusion] There are two typos in the conclusion: 'bootstraping' should be 'bootstrapping,' and 'compatability' should be 'compatibility.'

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central comparison is confounded by decoder and pipeline differences, but no derivation reduces to its own inputs by construction.

full rationale

Training minimizes an MSE reconstruction loss L = ||Y - V||_2^2 against the fixed CroCo feature map V, and evaluation uses ARI-FG against ground-truth MOVi masks via argmax over slot attention scores; neither the training objective nor the success metric is defined in terms of the paper's own outputs or fitted parameters. The claim that the geometric CroCo target causes the +10% ARI-FG gain over DINOSAUR in Table II is not identifiable from the reported comparisons, because the methods differ jointly in prediction target, decoder architecture, and temporal-slot pipeline (DINOSAUR is adapted from its image version with its own decoder, while the proposed method uses the attentional slot decoder and recurrent slot initialization). This is an experimental-control limitation, not a circular derivation: the reported numbers are not forced by construction, and the evaluation targets are external to the training signal. The paper's own observation in Section III-A that CroCo attention maps are diffuse and show no object grouping further indicates the claimed mechanism is not guaranteed by the representation choice. The only author self-citation is a background review reference and is not load-bearing. No uniqueness theorem is imported, no fitted parameter is renamed as a prediction, and no ansatz is smuggled in via citation.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No parameters are fitted to the evaluation metric; the only listed free parameter is the slot count, a standard hyperparameter. The central empirical comparison rests on the untested assumption that decoder and pipeline differences do not confound the representation-target comparison. No new physical entities, forces, or dimensions are introduced; the attentional slot decoder is a module, not an invented entity.

free parameters (1)

Number of object slots K = 11 for MOVi-A/C, 24 for MOVi-E
Chosen by hand to exceed the maximum object count in each dataset. It affects segmentation granularity and capacity but is not fitted to the ARI-FG metric. Standard hyperparameter, not a fitted model parameter.

assumptions (4)

domain assumption CroCo's pre-trained features encode geometric (3D) scene information that is useful for object-centric grouping.
Hypothesized in Section III-A based on CroCo's cross-view completion pretraining. The paper's quantitative comparison is the only evidence, and that comparison is confounded by decoder differences, so the benefit of geometric information is not independently demonstrated.
domain assumption The ARI-FG differences in Table II isolate the effect of the prediction target (geometric versus semantic features) across methods.
The methods use different decoders (our attentional decoder, DINOSAUR's slot-wise decoder, SA Vi's spatial broadcast decoder) and different training pipelines. The paper assumes these differences do not drive the gains. This is stated implicitly in Section IV-B where the improvement is attributed to the geometric target.
domain assumption MSE reconstruction of the pre-trained feature map V is a sufficient training signal for object-centric binding.
The method sets L = ||Y - V||_2^2 in Section III-C and relies on the pretrained features preserving object-level structure. The necessity or sufficiency of this reconstruction objective is not analyzed.
standard math Standard deep learning components (Vision Transformer, Slot Attention, LayerNorm, cross-attention, Adam) behave as documented in the cited literature.
The architecture relies on [7], [5], [22], and [14] for pretraining and core modules. These are accepted background results.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Efficient Object-centric Representation Learning with Pre-trained Geometric Prior." pith.science (2026). https://pith.science/paper/ETFWU7BQ

@misc{pith2026241212331,
  author       = {Pith},
  title        = {Pith review of: Efficient Object-centric Representation Learning with Pre-trained Geometric Prior},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ETFWU7BQ}},
  note         = {Machine review of arXiv:2412.12331}
}

read the original abstract

This paper addresses key challenges in object-centric representation learning of video. While existing approaches struggle with complex scenes, we propose a novel weakly-supervised framework that emphasises geometric understanding and leverages pre-trained vision models to enhance object discovery. Our method introduces an efficient slot decoder specifically designed for object-centric learning, enabling effective representation of multi-object scenes without requiring explicit depth information. Results on synthetic video benchmarks with increasing complexity in terms of objects and their movement, object occlusion and camera motion demonstrate that our approach achieves comparable performance to supervised methods while maintaining computational efficiency. This advances the field towards more practical applications in complex real-world scenarios.

Figures

Figures reproduced from arXiv: 2412.12331 by the authors.

**Figure 2.** Our overall architecture and training pipeline including the reconstruc [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Indicative qualitative results of our Attentional Decoder on the Movi [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Two examples of unsupervised object segmentation with our method [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 21 canonical work pages

[1]

On the Binding Problem in Artificial Neural Networks,

Klaus Greff, Sjoerd van Steenkiste, and J ¨urgen Schmidhuber, “On the Binding Problem in Artificial Neural Networks,” arXiv:2012.05208 [cs], Dec. 2020

arXiv 2012
[2]

Learn- ing object-centric transformation for video prediction,

Xiongtao Chen, Wenmin Wang, Jinzhuo Wang, and Weimian Li, “Learn- ing object-centric transformation for video prediction,” in Proc. 25th ACM International Conference on Multimedia , 2017, pp. 1503–1512

work page 2017
[3]

Attention over learned object embeddings enables complex visual reasoning,

David Ding, Felix Hill, Adam Santoro, Malcolm Reynolds, and Matt Botvinick, “Attention over learned object embeddings enables complex visual reasoning,” in Advances in Neural Information Processing Systems, 2021, vol. 34

work page 2021
[4]

Contrastive Learning of Structured World Models,

Thomas Kipf, Elise van der Pol, and Max Welling, “Contrastive Learning of Structured World Models,” in International Conference on Learning Representations , Sept. 2019

work page 2019
[5]

Object-Centric Learning with Slot Attention.,

Francesco Locatello, Dirk Weissenborn, Thomas Unterthiner, Aravindh Mahendran, Georg Heigold, Jakob Uszkoreit, Alexey Dosovitskiy, and Thomas Kipf, “Object-Centric Learning with Slot Attention.,” in Conference on Neural Information Processing Systems (NeurIPS) , 2020

work page 2020
[6]

Conditional Object-Centric Learning from Video,

Thomas Kipf, Gamaleldin F. Elsayed, Aravindh Mahendran, Austin Stone, Sara Sabour, Georg Heigold, Rico Jonschkowski, Alexey Doso- vitskiy, and Klaus Greff, “Conditional Object-Centric Learning from Video,” arXiv:2111.12594 [cs, stat] , Nov. 2021

arXiv 2021
[7]

An image is worth 16x16 words: Transformers for image recognition at scale,

Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weis- senborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby, “An image is worth 16x16 words: Transformers for image recognition at scale,” in 9th International Conference on Learning Representations, ICLR, 2021

work page 2021
[8]

Masked autoencoders are scalable vision learners,

Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll ´ar, and Ross Girshick, “Masked autoencoders are scalable vision learners,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 16000–16009

work page 2022

Show all 23 references

[9]

The effectiveness of MAE pre-pretraining for billion-scale pretraining.,

Mannat Singh, Quentin Duval, Kalyan Vasudev Alwala, Haoqi Fan, Vaibhav Aggarwal, Aaron Adcock, Armand Joulin, Piotr Doll ´ar, Christoph Feichtenhofer, Ross B. Girshick, Rohit Girdhar, and Ishan Misra, “The effectiveness of MAE pre-pretraining for billion-scale pretraining.,” i...

2023
[10]

Contrastive Representation Learning: A Framework and Review,

Phuc H. Le-Khac, Graham Healy, and Alan F. Smeaton, “Contrastive Representation Learning: A Framework and Review,” IEEE Access, vol. 8, pp. 193907–193934, Oct. 2020

2020
[11]

Bootstrap your own latent-a new approach to self-supervised learning,

Jean-Bastien Grill, Florian Strub, Florent Altch ´e, Corentin Tallec, Pierre Richemond, Elena Buchatskaya, Carl Doersch, Bernardo Avila Pires, Zhaohan Guo, and Mohammad Gheshlaghi Azar, “Bootstrap your own latent-a new approach to self-supervised learning,” Advances in neural ...

2020
[12]

Emerging properties in self-supervised vision transformers,

Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J ´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin, “Emerging properties in self-supervised vision transformers,” ICCV, 2021

2021
[13]

Omnimae: Single model masked pretraining on images and videos,

Rohit Girdhar, Alaaeldin El-Nouby, Mannat Singh, Kalyan Vasudev Alwala, Armand Joulin, and Ishan Misra, “Omnimae: Single model masked pretraining on images and videos,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 10406–10417

2023
[14]

CroCo: Self-supervised pre- training for 3D vision tasks by cross-view completion,

Philippe Weinzaepfel, Vincent Leroy, Thomas Lucas, Romain Br ´egier, Yohann Cabon, Vaibhav Arora, Leonid Antsfeld, Boris Chidlovskii, Gabriela Csurka, and J ´erˆome Revaud, “CroCo: Self-supervised pre- training for 3D vision tasks by cross-view completion,” in Annual Conferenc...

2022
[15]

Neural expectation maximization,

Klaus Greff, Sjoerd Van Steenkiste, and J ¨urgen Schmidhuber, “Neural expectation maximization,” Advances in Neural Information Processing Systems, vol. 30, 2017

2017
[16]

SIMONe: View-invariant, temporally-abstracted object repre- sentations via unsupervised video decomposition,

Rishabh Kabra, Daniel Zoran, Goker Erdogan, Loic Matthey, Antonia Creswell, Matt M. Botvinick, Alexander Lerchner, and Christopher P. Burgess, “SIMONe: View-invariant, temporally-abstracted object repre- sentations via unsupervised video decomposition,” in Annual Conference on...

2021
[17]

SA Vi++: Towards end-to-end object-centric learning from real-world videos,

Gamaleldin F. Elsayed, Aravindh Mahendran, Sjoerd van Steenkiste, Klaus Greff, Michael C. Mozer, and Thomas Kipf, “SA Vi++: Towards end-to-end object-centric learning from real-world videos,” in Annual Conference on Neural Information Processing Systems NeurIPS , 2022

2022
[18]

Bridging the Gap to Real-World Object-Centric Learning.,

Maximilian Seitzer, Max Horn, Andrii Zadaianchuk, Dominik Zietlow, Tianjun Xiao, Carl-Johann Simon-Gabriel, Tong He, Zheng Zhang, Bernhard Sch ¨olkopf, Thomas Brox, and Francesco Locatello, “Bridging the Gap to Real-World Object-Centric Learning.,” in International Conference ...

2023
[19]

Depth-aided Camouflaged Object Detection,

Qingwei Wang, Jinyu Yang, Xiaosheng Yu, Fangyi Wang, Peng Chen, and Feng Zheng, “Depth-aided Camouflaged Object Detection,” in Proceedings of the 31st ACM International Conference on Multimedia , 2023, pp. 3297–3306

2023
[20]

Kubric: A scalable dataset generator.,

Klaus Greff, Francois Belletti, Lucas Beyer, Carl Doersch, Yilun Du, Daniel Duckworth, David J. Fleet, Dan Gnanapragasam, Florian Golemo, Charles Herrmann, Thomas Kipf, Abhijit Kundu, Dmitry Lagun, Issam H. Laradji, Hsueh-Ti Derek Liu, Henning Meyer, Yishu Miao, Derek Nowrouze...

2022
[21]

Masked Siamese Networks for Label-Efficient Learning.,

Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Flo- rian Bordes, Pascal Vincent, Armand Joulin, Mike Rabbat, and Nicolas Ballas, “Masked Siamese Networks for Label-Efficient Learning.,” in European Conference on Computer Vision (ECCV) , 2022, pp. 456–473

2022
[22]

Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains.,

Matthew Tancik, Pratul P. Srinivasan, Ben Mildenhall, Sara Fridovich- Keil, Nithin Raghavan, Utkarsh Singhal, Ravi Ramamoorthi, Jonathan T. Barron, and Ren Ng, “Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains.,” in Conference on Neural I...

2020
[23]

Multi-Object Representation Learning with Iterative Variational Inference,

Klaus Greff, Rapha ¨el Lopez Kaufman, Rishabh Kabra, Nick Watters, Christopher Burgess, Daniel Zoran, Loic Matthey, Matthew Botvinick, and Alexander Lerchner, “Multi-Object Representation Learning with Iterative Variational Inference,” in International Conference on Machine Le...

2019

Pith tools

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

[1] [1]

On the Binding Problem in Artificial Neural Networks,

Klaus Greff, Sjoerd van Steenkiste, and J ¨urgen Schmidhuber, “On the Binding Problem in Artificial Neural Networks,” arXiv:2012.05208 [cs], Dec. 2020

arXiv 2012

[2] [2]

Learn- ing object-centric transformation for video prediction,

Xiongtao Chen, Wenmin Wang, Jinzhuo Wang, and Weimian Li, “Learn- ing object-centric transformation for video prediction,” in Proc. 25th ACM International Conference on Multimedia , 2017, pp. 1503–1512

work page 2017

[3] [3]

Attention over learned object embeddings enables complex visual reasoning,

David Ding, Felix Hill, Adam Santoro, Malcolm Reynolds, and Matt Botvinick, “Attention over learned object embeddings enables complex visual reasoning,” in Advances in Neural Information Processing Systems, 2021, vol. 34

work page 2021

[4] [4]

Contrastive Learning of Structured World Models,

Thomas Kipf, Elise van der Pol, and Max Welling, “Contrastive Learning of Structured World Models,” in International Conference on Learning Representations , Sept. 2019

work page 2019

[5] [5]

Object-Centric Learning with Slot Attention.,

Francesco Locatello, Dirk Weissenborn, Thomas Unterthiner, Aravindh Mahendran, Georg Heigold, Jakob Uszkoreit, Alexey Dosovitskiy, and Thomas Kipf, “Object-Centric Learning with Slot Attention.,” in Conference on Neural Information Processing Systems (NeurIPS) , 2020

work page 2020

[6] [6]

Conditional Object-Centric Learning from Video,

Thomas Kipf, Gamaleldin F. Elsayed, Aravindh Mahendran, Austin Stone, Sara Sabour, Georg Heigold, Rico Jonschkowski, Alexey Doso- vitskiy, and Klaus Greff, “Conditional Object-Centric Learning from Video,” arXiv:2111.12594 [cs, stat] , Nov. 2021

arXiv 2021

[7] [7]

An image is worth 16x16 words: Transformers for image recognition at scale,

Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weis- senborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby, “An image is worth 16x16 words: Transformers for image recognition at scale,” in 9th International Conference on Learning Representations, ICLR, 2021

work page 2021

[8] [8]

Masked autoencoders are scalable vision learners,

Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll ´ar, and Ross Girshick, “Masked autoencoders are scalable vision learners,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 16000–16009

work page 2022