Tuesday March 23 - Friday March 26, 2021
Reception: Tuesday evening
Presentations: Wednesday - Friday
Snowbird, Utah was the original planned physical location; however:
DCC 2021 will be 100% Virtual.
- University of Arizona and Brandeis University
- Microsoft Research
- IEEE Signal Processing Society (technical cosponsorship)
Proceedings published by
IEEE Computer Society Press CPS Online.
- Michael W. Marcellin, University of Arizona (DCC Co-Chair)
- James A. Storer, Brandeis University (DCC Co-Chair)
- Ali Bilgin, University of Arizona (Committee Co-Chair)
- Joan Serra-Sagrista, U. Autonoma de Barcelona (Committee Co-Chair)
- Henrique Malvar, Microsoft Research (Publications Chair)
- James E. Fowler, Mississippi State University (Publicity Chair)
- Johannes Ballé, Google
- Charles D. Creusere, New Mexico State University
- Travis Gagie, Dalhousie University
- Simon Gog, EBay
- Hamid Jafarkhani, University of California Irvine
- Giovanni Motta, Google, Inc.
- Gonzalo Navarro, University of Chile
- Yakov Nekrich, Michigan Technological University
- Jan Østergaard, Aalborg University
- Majid Rabbani, Rochester Institute of Technology
- Yuriy Reznik, Brightcove, Inc.
- Thomas Richter, Fraunhofer IIS
- Victor Sanchez, University of Warwick
- Serap Savari, Texas A&M University
- Khalid Sayood, University of Nebraska
- Peter Schelkens, Vrijie Universiteit Brussel
- Rahul Shah, Louisiana State University
- Dana Shapira, Ariel University
- Ofer Shayevitz, Tel Aviv University
- Gary J. Sullivan, Microsoft Corporation
- Aaron B. Wagner, Cornell University
- Jiangtao Wen, Tsinghua University
- Jizheng Xu, Bytedance Inc.
- En-Hui Yang, University of Waterloo
- Yan Ye, Alibaba Group
- Peng Yin, Dolby Laboratories, Inc.
An international forum for current work on data compression and related applications.
Both theoretical and experimental work are of interest.
Topics of interest include but are not limited to:
Lossless and lossy compression for storage and transmission of specific types of data
gray scale and color photographs,
multi-spectral and hyper-spectral images,
instrument and sensor data,
earth observation data,
bi-level images / bit-maps,
web graphs, etc.),
source coding in multiple access networks,
joint source-channel coding,
rate distortion coding,
multiple description coding,
vector quantization (VQ),
multiple description VQ,
transform based methods (including DCT and wavelet transforms),
parallel compression algorithms and hardware,
error resilient compression techniques,
adaptive compression algorithms,
browsing and searching compressed data,
compressed data structures,
applications to immersive media,
applications of neural networks and deep learning (e.g. CNN's) to compression,
string searching and manipulation used in compression applications,
fractal based compression methods,
information retrieval employing compression techniques,
steganography / hidden information with respect to compressed data,
minimal length encoding and applications to learning,
system issues relating to data compression
(including error control, data security, indexing, and browsing),
compression applications and issues for computational biology and bioinformatics,
compression applications and issues for the internet,
compression applications and issues for mobile computing,
applications of compression to file distribution and software updates,
applications of compression to file storage and backup systems,
applications of compression to data mining,
applications of compression to image retrieval,
applications of compression and information theory to human-computer interaction (HCI),
development of and extensions to compression standards
(including the HEVC, JPEG, MPEG, H.xxx, and G.xxx families
and including compression of specific image types
such as plenoptic images, point cloud images, and light field images),
compressed sensing / compressive sampling,
the use of techniques from information theory and data compression in
networking, communications, and storage of large data sets.
"Quality is in the Eye of the Beholder"
Prof. Alan Bovik
The University of Texas at Austin
In addition to general sessions addressing
the theme described above,
submissions are being sought for these special sessions:
"Video Coding Technologies"
Gary Sullivan, Microsoft Research
Video coding technologies for compression performance improvement
over existing codecs (VVC, HEVC, AV1, etc.) based on the hybrid
video coding framework or other frameworks such as those based on
deep-learning technologies, video coding technologies for
emerging applications (immersive video, omni-directional video,
AR, VR and XR, etc.) and/or with enhanced functionalities (BEAM,
tiles, subpictures, ultra-low latency streaming, adaptive
streaming etc.), encoder optimization algorithms for encoder
complexity reduction and/or compression performance improvement
using conventional signal processing techniques and deep-learning
based techniques, software and hardware implementation of various
video codecs (such as HEVC, VVC, AV1, etc.), quality measurement
and optimization of video coding systems (including design of
subjective and objective quality metrics and video pre- and
post-processing algorithms), and performance measurement and
comparison of various video codecs (such as VVC, HEVC, AV1, etc.).
Yan Ye, Alibaba Group
Yuriy Reznik, Brightcove, Inc.
"Compression and Quantization in Learning"
Hamid Jafarkhani, University of California Irvine
Compressing different components during the learning process,
distributed learning, federated learning, quantized learning,
quantization-aware training, quantized neural networks, using
compression algorithms in machine learning tasks such as clustering and
classification, and learning with quantized and sub-Nyquist sampled data.
Jiangtao Wen, Tsinghua University
"Neural Networks for Compression"
Aaron Wagner, Cornell University
Neural networks, especially deep neural networks,
have shown tremendous promise as compressors, especially for
multimedia sources. This session will spotlight the latest
advances in neural-network based-compressors. We invite
submissions on all aspects of neural-network-based compression,
with a particular emphasis on (1) improved methods for
structuring and training neural networks for compression, (2)
methods for combining neural networks with classical compression
techniques, and (3) the identification of new applications for
which neural-network based compressors are particularly well
suited. Authors are requested to make transparent by which means
the proposed network solves the compression problem and which
lessons can be learned from their contribution. Compression of
trained neural network models themselves is outside the scope of
this session but within the scope of the conference at large.
Submissions will be judged with an eye toward novelty,
reproducibility, generalizability, and coherence of the session.
Johannes Ballé, Google
Thomas Richter, Fraunhofer IIS
All submitted manuscripts must be PDF files that satisfy:
Each page should have a top margin of 1 inch and a left margin
of 1.25 inches, and the text area on each page should be 9 inches high by 6
Do NOT use a two-column format; all submissions must have a single column format.
Use 12 point type.
The first page must begin with the title centered at the top,
with the authors and their affiliations below the title.
(DCC reviewing is
where the identities of the reviewers are anonymous,
but the author names and affiliations are on the manuscript.)
Following the title and authors,
should be a short abstract,
and following the abstract should be
the start of the first section.
Manuscripts may NOT be more than 10 pages TOTAL,
including all references, figures, tables, notes, and appendices.
In the case of an image or video that requires viewing to access visual quality,
for optional use by a referee,
a reduced size but reasonably viewable image or representative video frame
may appear in the manuscript along with a link to the image or the video.
In the case of audio,
for optional use by a referee,
a description of the audio material may be accompanied by a link to the audio.
Here is a link to template of a formatted page:
DCC Template (pdf)
Submissions must be in PDF format, but here are some files that may be useful for preparation:
MS Word sample source file (.doc)
LaTeX sample source files (zip folder)
Manuscripts may be submitted for consideration as full
paper or poster, or for consideration as poster only. Manuscripts submitted
for consideration as poster only must still submit a thorough description of
the work for review (NOT a one page summary).
Manuscripts accepted as papers will be presented at a
technical session of the conference and have a final draft of at most 10
pages in the DCC proceedings. Manuscripts accepted as posters will be
presented at the DCC poster session and have a one page summary in the DCC
For inclusion in the DCC proceedings,
an accepted manuscript requires the same formatting as for submission,
the only difference being that manuscripts accepted as papers are limited to 10 pages
and manuscripts accepted as posters are limited to one page.
Submissions must be submitted electronically;
November 9, 11:59pm U.S. Pacific Time.
(This date is an extension by one week from the original deadline of Nov. 2.)
DCC observes the
on confidentiality of submitted manuscripts.
Click here to go to the DCC submissions page.
Submissions may be posted on a preprint server provided
the IEEE posting policy is followed:
IEEE policy page on posting your manuscript
Authors will be notified via email in late December of
acceptance as a paper, acceptance as a poster, or rejection. Accepted
manuscripts must be submitted electronically; the due date will be in early
January. The letter of acceptance will include the exact due date, directions
on where and how to make the electronic submission of the final manuscript,
and directions for submitting a proceedings copyright form. (Do not include a
copyright form with your submission; instead, wait until your submission has been
accepted and you receive directions.)
DCC Call for Papers (PDF)