Pith. sign in

REVIEW 3 major objections 4 minor 53 references

DSIC: Deep Stereo Image Compression

T0 review · 3 major / 4 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 A stereo pair can be compressed as one joint code, with the second image costing 30–50% fewer bits at low bitrates and up to 29% less joint bitrate, at equal or better quality.

desk verdict First credible end-to-end learned stereo image codec with a genuinely novel architecture, but the headline bitrate savings are partly a comparison artifact from the entropy-model upgrade and the second-image numbers are inferred, not directly measured. read the letter →

arxiv 1908.03631 v1 pith:TPSJQHXA submitted 2019-08-09 eess.IV cs.CV

Jerry Liu , Shenlong Wang , Raquel Urtasun This is my paper · ORCID
classification eess.IVcs.CV
keywords stereoimagecompressionlearneddisparitywarpingconditionalentropymodelparametricskipfunctionrate-distortionoptimizationMS-SSIMautonomousdrivingdata
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 claims that a single learned model can compress a rectified stereo pair more cheaply than compressing the two images independently, by actively reusing the first image's content when coding the second. The model adds parametric skip functions that estimate a per-level disparity cost volume, densely warp the first image's features into the second image's encoder and decoder, and condition the second image's latent-code entropy on the first image's code. Across two large driving datasets, the authors report a 29.0% reduction in joint bitrate at low bitrates and a 17.8% reduction at higher bitrates on NorthAmerica, at equal or better MS-SSIM/PSNR than a single-image hyperprior codec, with 30–50% savings in the second image's residual bitrate at low bitrates. If this holds, overlapping cameras need not store redundant bits, which matters for bandwidth-constrained settings such as self-driving fleets.

What carries the argument

The load-bearing object is the parametric skip function combined with a conditional entropy model. The skip function takes the previous feature maps from both images plus a global context read out of the first image's code, forms a $C \times H \times W$ cost volume normalized by softmax along the disparity axis ($C$ is the hand-set maximum disparity: 32 on NorthAmerica, 64 on Cityscapes), and produces the aligned feature via $g_{2,i} = \sum_{d=0}^{C} c_{d,i}\,h_{1,(i,d)}$, turning disparity estimation into a differentiable attention-like weighting rather than a single hard shift. The conditional entropy model then codes the second quantized latent against the first latent and a hyperprior using a Gaussian-mixture density, capturing the residual correlation that warping leaves behind. Together these two mechanisms carry the claimed bitrate savings.

What would settle it

Run the same architecture on rectified pairs with known ground-truth disparity while raising the fraction of pixels whose disparity exceeds $C$ (for example by placing objects very close to the camera), and compare joint bitrate against the same model with $C$ doubled or against a version with $C=0$; if the savings do not drop when true disparities exceed $C$, the disparity range is not the limiting assumption. A simpler variant: synthetically shift one image by a constant offset larger than $C$ and check whether the second-image bitrate returns to single-image levels.

Watch

Extended reading notes

Core claim

The central discovery is that inter-image redundancy in a stereo pair can be exploited end-to-end by two coupled mechanisms. First, a parametric skip function predicts, at each encoder/decoder level, a soft cost volume over a fixed disparity range and uses it to densely warp the first image's feature map toward the second: the warped feature at pixel $i$ is a weighted sum over disparities $d=0..C$ of the first image's features. Second, a conditional entropy model factors the joint bitstream as $p(\bar{y}_2|\bar{y}_1,\bar{z}_2)\,p(\bar{y}_1|\bar{z}_1)\,p(\bar{z}_2)\,p(\bar{z}_1)$, so the second image's code is coded relative to the first. The paper argues that this joint model lowers the total bitrate below independent coding while keeping or improving reconstruction quality, and supports it with comparisons against image codecs, video codecs, and a stereo residual-coding baseline.

Load-bearing premise

The bitrate savings rest on the assumption that the two images are rectified and that every transferable pixel's true disparity lies within the preset range $0..C$ (32 or 64 pixels); where that fails, the skip functions cannot pull the needed information across and the savings shrink toward single-image coding.

Editorial extensions

If this is right

  • Storing a stereo pair jointly can roughly double the per-image savings relative to independent compression, so the meaningful headline number is the residual bitrate reduction for the second image.
  • At low bitrates the second image costs 30–50% fewer bits than single-image coding, which makes the method most attractive for bandwidth-limited applications; at high bitrates the advantage narrows to 10–20%.
  • The model needs no ground-truth disparity supervision: the cost volumes emerge from the rate-distortion objective, so the same architecture can be trained on any rectified stereo corpus.
  • In ablations, disparity warping alone helps most at low bitrates, while the conditional entropy model helps across all bitrates; combining them beats either alone, though some of the gain is shared.
  • Against MS-SSIM the joint model surpasses conventional codecs and motion-compensation-plus-residual baselines; on Cityscapes PSNR, BPG and MV-HEVC remain competitive, so the advantage is clearest on perceptual structure rather than raw MSE.

Reading between the lines

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

  • The fixed disparity range $C$ is the method's implicit geometry prior; on scenes with large close-up disparities (e.g., pedestrians right beside the camera) the savings should shrink, and a systematic study of bitrate savings vs. baseline width or scene depth would map that boundary.
  • The same machinery could be lifted to video by replacing the disparity axis with optical flow or time-varying displacement, since the cost volume and conditional entropy do not depend on the two frames being from a stereo camera.
  • Because warp-based transfer and conditional entropy both consume the same inter-image correlation, their partial redundancy (which the paper observes as cannibalization) suggests a cleaner decomposition into shared vs. unique information could yield further gains.
  • The paper reports a large compute cost (roughly four times the FLOPs of the single-image model, and 141 GFLOPS at 480x300), so a practical deployment would need sparse cost volumes or learned disparity priors; testing such approximations would show how much of the bitrate gain survives.
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

3 major / 4 minor

Summary. The paper proposes DSIC, a deep stereo image compression architecture that compresses a rectified stereo pair jointly. The first image is encoded with a single-image autoencoder; the second image stream receives multi-level, disparity-warped features from the first image through parametric skip functions, and a conditional entropy model (a Gaussian mixture model) models the second code given the first. The whole system is trained end-to-end with a rate-distortion loss. Experiments on a proprietary NorthAmerica driving dataset and on Cityscapes compare against Ballé's hyperprior model, BPG, JPEG2000, HEVC/MV-HEVC, and a stereo residual-coding baseline, reporting joint bitrate savings, inferred second-image savings, and ablations over skip functions, conditional entropy, and hyperpriors.

Significance. If the empirical claims hold, the paper would be an early demonstration that learned multi-image compression can exploit cross-view redundancy beyond single-image codecs, with real coded bitrates verified within 1-2% of the entropy estimate (Appendix D). The architecture is clearly described, the end-to-end training objective is standard, and the ablation study attempts to separate the contributions of the skip functions and the conditional entropy model. The main significance depends on whether the comparison against the single-image baseline isolates the stereo-specific gain, which is currently not established.

major comments (3)
  1. [Section 3.3, Eq. (14), Appendix C.3, Section 4.3] The comparison to Ballé's hyperprior baseline does not isolate the stereo-specific gain from an entropy-model capacity upgrade. In DSIC, both image codes are modeled with a Gaussian mixture model (Eqs. (12)-(14)), while the Ballé baseline uses a single Gaussian conditional density; a mixture distribution strictly has lower cross-entropy for the same latent codes. Therefore part of the reported 29% joint-bitrate reduction and the inferred 58% residual-bitrate reduction may come from the GMM entropy model for the first image, not from cross-image coding. The ablation in Fig. 9 does not include a control where the first image uses the same GMM-based hyperprior entropy model but no skip functions and no conditioning of the second code on the first. Please add such a control and report the rate-distortion curve against the full stereo model.
  2. [Section 4.3, Appendix A] The headline 'second image bitrate' savings are not directly measured. The paper reports joint bitrate savings and then assumes 'our stereo pair is compressed as well under our model as under a single-image setting' to double the joint savings into residual savings. This assumption is not validated and is in tension with Appendix A, which states that the first code may contain additional information to help compress the second code. The first image may therefore carry extra bits relative to a pure single-image code, and the second image may carry fewer, so the decomposition 'second-image savings = 2 × joint savings' is an extrapolation, not an observed quantity. Please report actual per-image bitrates for both images under the stereo model and under a matched single-image model, or revise the abstract and conclusion to state joint-bitrate savings without the inferred residual percentages.
  3. [Eq. (3), Section 4.2] The approach relies on a rectified-stereo assumption and a hand-fixed maximum disparity C (32 for NorthAmerica, 64 for Cityscapes, Section 4.2). If a pixel's true disparity exceeds C or the input is not rectified, the warp in Eq. (3) cannot transfer the needed information and the claimed savings degrade. The paper does not report any sensitivity analysis to C, nor does it state the range of disparities present in the test sets. Please add an experiment that varies C for a fixed model (or reports the fraction of disparity estimates at the boundary of the range) to bound the regime in which the claims hold.
minor comments (4)
  1. [Section 4.3] The sentence 'At a higher Ballé bitrate 0.734, our model achieves a 17.8% reduction in the residual' appears inconsistent with the preceding sentence, which reports joint-bitrate reductions; please clarify whether 17.8% is the joint or the residual reduction.
  2. [Section 5] There are typos: 'parameteric' should be 'parametric', and 'demostrate' should be 'demonstrate'.
  3. [Eq. (3)] The notation '(i,d) represents the pixel index that is d pixels right of pixel i' is ambiguous about the direction convention for disparity on rectified stereo pairs; please specify the sign convention for d.
  4. [Table 1] Table 1 varies resolution and camera baseline simultaneously, so the individual effect of baseline width cannot be separated from resolution; please state the intended confound or provide a version that holds resolution fixed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the stereo bitrate claims are measured with real range-coded bitstreams against an external single-image baseline, and the GMM-vs-Gaussian entropy-model mismatch is a baseline-fairness issue, not a circular reduction.

full rationale

Walking the claimed derivation chain, the stereo codec is trained end-to-end on the rate-distortion objective of Eq. (15), where the rate is the cross-entropy under the learned conditional entropy model of Eqs. (5)-(14). This is standard learned-compression practice, and the paper does not stop at the training objective: Appendix D reports that its range coder encodes within 1-2% of the Shannon entropy bound, so the reported bits are actual coded bitstream lengths rather than fitted values relabeled as predictions. The central comparison in Section 4.3 is against Ballé et al. [5], an external single-image baseline, and the ablation studies in Figs. 5 and 9 isolate the contributions of DispSkip, CE0, and the hyperprior rather than asserting the stereo gain directly from the baseline. The only author-affiliated citation is [28], used as background on stereo matching, and it bears no weight in the rate-distortion derivation; no uniqueness theorem or ansatz is imported from the authors' prior work. The Section 4.3 statement that residual savings are double the joint savings rests on a stated assumption about the first-image bitrate; if the GMM entropy model lowers the first-image rate relative to Ballé's single-Gaussian hyperprior, the stereo-specific saving could be overstated, but that is a benchmark-fairness confound, not a circular argument. No equation in the paper reduces to its own input, and no fitted parameter is renamed as an independent prediction.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new physical entities; the parametric skip function and global context volume are neural network modules. The free parameters listed are hand-chosen training/architecture settings on which the reported savings depend.

free parameters (2)
  • max_disparity C = 32 (NorthAmerica), 64 (Cityscapes)
    Chosen by hand as the disparity search range for the cost volume; the warp in Eq. (3) only covers disparities 0..C, so the claimed bitrate savings depend on this choice covering the true disparities in the data.
  • rate-distortion weight beta = 0.5 to 0.001 for lower to higher bitrate models
    Set per target bitrate to sweep the rate-distortion curve; standard training hyperparameter, but the reported curves are defined by these hand-chosen values.
assumptions (3)
  • domain assumption The joint distribution factorizes as p(z1)p(z2)p(y1|z1)p(y2|y1,z2) (Eq. 5), i.e., y2 is conditionally independent of z1 given y1, and z1,z2 are independent.
    This factorization is a modeling choice for the entropy model; if the true dependence structure is richer, the estimated rate differs, though range coding would still produce correct lengths under the model.
  • domain assumption Input images are rectified stereo pairs with all correspondences within disparity range [0,C] (Eq. 3).
    The dense warp samples the first image's feature at positions i-d for d in [0,C]; out-of-range or non-rectified correspondences cannot be aligned.
  • domain assumption Each latent pixel's conditional distribution is a Gaussian mixture (Eq. 13) with parameters predicted by a neural network.
    The GMM is a flexible prior; the paper reports it helps slightly, but any approximation error carries into the rate estimate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DSIC: Deep Stereo Image Compression." pith.science (2026). https://pith.science/paper/TPSJQHXA

@misc{pith2026190803631,
  author       = {Pith},
  title        = {Pith review of: DSIC: Deep Stereo Image Compression},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TPSJQHXA}},
  note         = {Machine review of arXiv:1908.03631}
}
read the original abstract

In this paper we tackle the problem of stereo image compression, and leverage the fact that the two images have overlapping fields of view to further compress the representations. Our approach leverages state-of-the-art single-image compression autoencoders and enhances the compression with novel parametric skip functions to feed fully differentiable, disparity-warped features at all levels to the encoder/decoder of the second image. Moreover, we model the probabilistic dependence between the image codes using a conditional entropy model. Our experiments show an impressive 30 - 50% reduction in the second image bitrate at low bitrates compared to deep single-image compression, and a 10 - 20% reduction at higher bitrates.

Figures

Figures reproduced from arXiv: 1908.03631 by the authors.

Figure 1
Figure 1. Left: the end-to-end stereo compression architecture; right: the proposed deep parameteric skip function. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Visualization of the disparity volumes at each resolution [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Plot of our stereo model against various compression baselines, for NorthAmerica and Cityscapes. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Qualitative results of our method vs. various single-image baselines. Metrics are specified for each entry. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Ablation study. For both datasets, we analyze the independent and combined effects of our skip functions (DispSkip), conditional [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Comparison between the reconstructions of competing baselines and our method on a NorthAmerica stereo pair. We observe that [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 6
Figure 6. Figure 6: (Continued) bining DispSkip with conditional entropy, which is also ob￾served in the main paper; yet again, combining DispSkip with conditional entropy yields the best results at all bi￾trates. C. Additional Architecture Details We provide additional architecture detai…
Figure 7
Figure 7. Figure 7: Input/Output disparity map, residual image, and camera image for a sample stereo pair on NorthAmerica (using our stereo [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: Comparison between the reconstructions from the two cameras using our deep residual coding baseline as well as our stereo [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]
Figure 9
Figure 9. Figure 9: Additional ablation study. For both datasets, we analyze the independent and combined effects of our skip functions (DispSkip) [PITH_FULL_IMAGE:figures/full_fig_p016_9.png]
Figure 10
Figure 10. Figure 10: Architecture diagram for producing the global context volume from [PITH_FULL_IMAGE:figures/full_fig_p017_10.png]
Figure 11
Figure 11. Figure 11: Architecture diagram for producing the cost volume from [PITH_FULL_IMAGE:figures/full_fig_p017_11.png]
Figure 12
Figure 12. Figure 12: Architecture diagram detailing the hyper-encoder (left) as well as the full entropy model of [PITH_FULL_IMAGE:figures/full_fig_p019_12.png]
Figure 13
Figure 13. Figure 13: Architecture diagram illustrating how σ is predicted for y¯2. The key difference is that y¯1 is concatenated with an upsampled z¯2 and the convolutions are no longer upsampling. The changes to predict µ, w are the same. 4339 [PITH_FULL_IMAGE:figures/full_fig_p019_13.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 48 canonical work pages

  1. [1]

    Available at https://github.com/listenlink/3D-HEVC

    Mv-hevc implementation. Available at https://github.com/listenlink/3D-HEVC

    work page

  2. [2]

    Soft-to-hard vector quantization for end-to-end learn- ing compressible representations

    Eirikur Agustsson, Fabian Mentzer, Michael Tschannen, Lukas Cavigelli, Radu Timofte, Luca Benini, and Luc V Gool. Soft-to-hard vector quantization for end-to-end learn- ing compressible representations. In NeurIPS, 2017

    work page 2017

  3. [3]

    Haluk Aydinoglu and Monson H. Hayes. Stereo image cod- ing: a projection approach. IEEE Transactions on Image Processing, 1998

    work page 1998

  4. [4]

    Simoncelli

    Johannes Ball, Valero Laparra, and Eero P. Simoncelli. End- to-end optimized image compression. In ICLR, 2017. 4328

    work page 2017

  5. [5]

    Variational image compression with a scale hyperprior

    Johannes Ball, David Minnen, Saurabh Singh, Sung Jin Hwang, and Nick Johnston. Variational image compression with a scale hyperprior. In ICLR, 2018

    work page 2018

  6. [6]

    Simoncelli

    Johannes Ball ´e, Valero Laparra, and Eero P. Simoncelli. Density modeling of images using a generalized normaliza- tion transformation. In ArXiv, 2015

    work page 2015

  7. [7]

    PatchMatch: A randomized correspon- dence algorithm for structural image editing

    Connelly Barnes, Eli Shechtman, Adam Finkelstein, and Dan B Goldman. PatchMatch: A randomized correspon- dence algorithm for structural image editing. InSIGGRAPH, 2009

    work page 2009

  8. [8]

    Depth discontinuities by pixel-to-pixel stereo

    Stan Birchfield and Carlo Tomasi. Depth discontinuities by pixel-to-pixel stereo. In ICCV, 1998

    work page 1998

Show all 53 references
  1. [9]

    Pyramid stereo matching network

    Jia-Ren Chang and Yong-Sheng Chen. Pyramid stereo matching network. In CVPR, 2018

    2018

  2. [10]

    The cityscapes dataset for semantic urban scene understanding

    Marius Cordts, Mohamed Omran, Sebastian Ramos, Timo Rehfeld, Markus Enzweiler, Rodrigo Benenson, Uwe Franke, Stefan Roth, and Bernt Schiele. The cityscapes dataset for semantic urban scene understanding. In CVPR, 2016

    2016

  3. [11]

    Multiview video compres- sion

    Markus Flierl and Bernd Girod. Multiview video compres- sion. IEEE Signal Processing Magazine, 2007

    2007

  4. [12]

    Motion and disparity compensated coding for multiview video.IEEE Transactions on Circuits and Systems for Video Technology, 2007

    Markus Flierl, Aditya Mavlankar, and Bernd Girod. Motion and disparity compensated coding for multiview video.IEEE Transactions on Circuits and Systems for Video Technology, 2007

    2007

  5. [13]

    Deepstereo: Learning to predict new views from the world’s imagery

    John Flynn, Ivan Neulander, James Philbin, and Noah Snavely. Deepstereo: Learning to predict new views from the world’s imagery. InCVPR, 2016

    2016

  6. [14]

    Residual image coding for stereo image compression

    Tamas Frajka and Kenneth Zeger. Residual image coding for stereo image compression. In ICIP, 2002

    2002

  7. [15]

    Deep probabilistic video compression

    Jun Han, Salvator Lombardo, Christopher Schroers, and Stephan Mandt. Deep probabilistic video compression. In ArXiv, 2018

    2018

  8. [16]

    Computer Matching of Areas in Stereo Images

    Marsha Jo Hannah. Computer Matching of Areas in Stereo Images. PhD thesis, Stanford, CA, USA, 1974

    1974

  9. [17]

    Stereo processing by semiglobal matching and mutual information

    Heiko Hirschmuller. Stereo processing by semiglobal matching and mutual information. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008

    2008

  10. [18]

    Batch normalization: Accelerating deep network training by reducing internal co- variate shift

    Sergey Ioffe and Christian Szegedy. Batch normalization: Accelerating deep network training by reducing internal co- variate shift. In ICML, 2015

    2015

  11. [19]

    Spatial transformer networks

    Max Jaderberg, Karen Simonyan, Andrew Zisserman, and Koray Kavukcuoglu. Spatial transformer networks. In NIPS, 2015

    2015

  12. [20]

    Light field image compression using generative adversarial network based view synthesis

    Chuanmin Jia, Xinfeng Zhang, Shanshe Wang, Shiqi Wang, Shiliang Pu, and Siwei Ma. Light field image compression using generative adversarial network based view synthesis. IEEE Journal on Emerging and Selected Topics in Circuits and Systems, 2018

    2018

  13. [21]

    Light field compression using depth image based view syn- thesis

    Xiaoran Jiang, Mikael Le Pendu, and Christine Guillemot. Light field compression using depth image based view syn- thesis. In ICMEW, 2017

    2017

  14. [22]

    End-to-end learning of geometry and context for deep stereo regression

    Alex Kendall, Hayk Martirosyan, Saumitro Dasgupta, Peter Henry, Ryan Kennedy, Abraham Bachrach, and Adam Bry. End-to-end learning of geometry and context for deep stereo regression. In ICCV, 2017

    2017

  15. [23]

    Multi-view video coding using view interpolation and reference picture selection

    Masaki Kitahara, Hideaki Kimata, Shinya Shimizu, Kazuto Kamikura, Yoshiyuki Yashima, Kenji Yamamoto, Tomohiro Yendo, Toshiaki Fujii, and Masayuki Tanimoto. Multi-view video coding using view interpolation and reference picture selection. ICME, 2006

    2006

  16. [24]

    Context-adaptive entropy model for end-to-end optimized image compression

    Jooyoung Lee, Seunghyun Cho, and Seung-Kwon Beack. Context-adaptive entropy model for end-to-end optimized image compression. In ICLR, 2019

    2019

  17. [25]

    Neural Video Compression using Spatio-Temporal Priors

    Haojie Liu, Tong Chen, Ming Lu, Qiu Shen, and Zhan Ma. Neural Video Compression using Spatio-Temporal Priors. In ArXiv, 2019

    2019

  18. [26]

    Jean loup Gailly and Mark Adler. zlib. https://www. zlib.net/, 1995–2017

    1995

  19. [27]

    Michael E. Lukacs. Predictive coding of multi-viewpoint im- age sets. In ICASSP, 1986

    1986

  20. [28]

    Schwing, and Raquel Urtasun

    Wenjie Luo, Alexander G. Schwing, and Raquel Urtasun. Ef- ficient deep learning for stereo matching. CVPR, 2016

    2016

  21. [29]

    Range encoding: an algorithm for removing redundancy from a digitised message

    Gloria Mart ´ın. Range encoding: an algorithm for removing redundancy from a digitised message. In Video and Data Recording Conference, 1979

    1979

  22. [30]

    View synthesis for multiview video compression

    Emin Martinian, Alexander Behrens, Jun Xin, and Anthony Vetro. View synthesis for multiview video compression. In Picture Coding Symposium, 2006

    2006

  23. [31]

    Conditional probability models for deep image compression

    Fabian Mentzer, Eirikur Agustsson, Michael Tschannen, Radu Timofte, and Luc Van Gool. Conditional probability models for deep image compression. In CVPR, 2018

    2018

  24. [32]

    Efficient compression of multi-view video exploiting inter-view dependencies based on h.264/mpeg4- avc

    Philipp Merkle, Karsten Muller, Aljoscha Smolic, and Thomas Wiegand. Efficient compression of multi-view video exploiting inter-view dependencies based on h.264/mpeg4- avc. In ICME, 2006

    2006

  25. [33]

    Joint au- toregressive and hierarchical priors for learned image com- pression

    David Minnen, Johannes Ball, and George Toderici. Joint au- toregressive and hierarchical priors for learned image com- pression. In NIPS, 2018

    2018

  26. [34]

    Moellenhoff and Mark W

    Mark S. Moellenhoff and Mark W. Maier. Transform cod- ing of stereo image residuals. IEEE Transactions on Image Processing, 1998

    1998

  27. [35]

    Real-time adaptive im- age compression

    Oren Rippel and Lubomir Bourdev. Real-time adaptive im- age compression. In ICML, 2017

    2017

  28. [36]

    Anderson, and Lubomir Bourdev

    Oren Rippel, Sanjay Nair, Carissa Lew, Steve Branson, Alexander G. Anderson, and Lubomir Bourdev. Learned video compression. In ArXiv, 2018

    2018

  29. [37]

    Ryan, R.T

    Thomas W. Ryan, R.T. Gray, and Bobby R. Hunt. Prediction of correlation errors in stereo pair images. Optical Engineer- ing, 1980

    1980

  30. [38]

    Learning conditional ran- dom fields for stereo

    Daniel Scharstein and Chris Pal. Learning conditional ran- dom fields for stereo. In CVPR, 2007

    2007

  31. [39]

    Claude E. Shannon. A mathematical theory of communica- tion. Bell System Technical Journal, 1948

    1948

  32. [40]

    Mel Siegel, Priyan Gunatilake, Sriram Sethuraman, and An- gel G. Jordan. Compression of stereo image pairs and streams. In Stereoscopic Displays and Virtual Reality Sys- tems, 1994

    1994

  33. [41]

    Sullivan, Jens-Rainer Ohm, Woo-Jin Han, and Thomas Wiegand

    Gary J. Sullivan, Jens-Rainer Ohm, Woo-Jin Han, and Thomas Wiegand. Overview of the high efficiency video coding (hevc) standard. IEEE Transactions on Circuits and Systems for Video Technology, 2012. 4329

    2012

  34. [42]

    Stereoscopic coding system, 1997

    Katsumi Tahara. Stereoscopic coding system, 1997. US Patent 5,633,682

    1997

  35. [43]

    Lossy image compression with compressive autoen- coders

    Lucas Theis, Wenzhe Shi, Andrew Cunningham, and Ferenc Huszar. Lossy image compression with compressive autoen- coders. In ICLR, 2017

    2017

  36. [44]

    O’Malley, Sung Jin Hwang, and Damien Vincent

    George Toderici, Sean M. O’Malley, Sung Jin Hwang, and Damien Vincent. Variable rate image compression with re- current neural networks. In ICLR, 2016

    2016

  37. [45]

    Full resolution image compression with recurrent neural net- works

    George Toderici, Damien Vincent, Nick Johnston, Sung Jin Hwang, David Minnen, Joel Shor, and Michele Covell. Full resolution image compression with recurrent neural net- works. In CVPR, 2017

    2017

  38. [46]

    Simoncelli, and Alan C

    Zhou Wang, Eero P. Simoncelli, and Alan C. Bovik. Multi- scale structural similarity for image quality assessment. In ACSSC, 2003

    2003

  39. [47]

    Sullivan, Gisle Bjontegaard, and Ajay Luthra

    Thomas Wiegand, Gary J. Sullivan, Gisle Bjontegaard, and Ajay Luthra. Overview of the h.264/avc video coding stan- dard. IEEE Transactions on Circuits and Systems for Video Technology, 2003

    2003

  40. [48]

    Stereo image compression with disparity compensation using the mrf model

    Antonio Ortega Woontack Woo. Stereo image compression with disparity compensation using the mrf model. 1996

    1996

  41. [49]

    Video compression through image interpolation

    Chao-Yuan Wu, Nayan Singhal, and Philipp Krahenbuhl. Video compression through image interpolation. In ECCV, 2018

    2018

  42. [50]

    Group normalization

    Yuxin Wu and Kaiming He. Group normalization. In ECCV, 2018

    2018

  43. [51]

    Deep3d: Fully automatic 2d-to-3d video conversion with deep con- volutional neural networks

    Junyuan Xie, Ross Girshick, and Ali Farhadi. Deep3d: Fully automatic 2d-to-3d video conversion with deep con- volutional neural networks. In ECCV, 2016

    2016

  44. [52]

    Stereo matching by training a convolutional neural network to compare image patches

    Jure Zbontar and Yann LeCun. Stereo matching by training a convolutional neural network to compare image patches. JMLR, 2016

    2016

  45. [53]

    Self- supervised learning for stereo matching with self-improving ability

    Yiran Zhong, Yuchao Dai, and Hongdong Li. Self- supervised learning for stereo matching with self-improving ability. In CVPR, 2017. 4330 A. Additional Qualitative Results We showcase additional qualitative results comparing our stereo compression model against other baselines,...

    2017

Pith tools

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