Internet DRAFT - draft-valin-videocodec-pvq
draft-valin-videocodec-pvq
Network Working Group JM. Valin
Internet-Draft Mozilla
Intended status: Standards Track March 9, 2015
Expires: September 10, 2015
Pyramid Vector Quantization for Video Coding
draft-valin-videocodec-pvq-02
Abstract
This proposes applying pyramid vector quantization (PVQ) to video
coding.
Status of This Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress."
This Internet-Draft will expire on September 10, 2015.
Copyright Notice
Copyright (c) 2015 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as
described in the Simplified BSD License.
This document may not be modified, and derivative works of it may not
be created, and it may not be published except as an Internet-Draft.
Valin Expires September 10, 2015 [Page 1]
Internet-Draft Video PVQ March 2015
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Gain-Shape Coding and Activity Masking . . . . . . . . . . . 2
3. Householder Reflection . . . . . . . . . . . . . . . . . . . 3
4. Angle-Based Encoding . . . . . . . . . . . . . . . . . . . . 4
5. Bi-prediction . . . . . . . . . . . . . . . . . . . . . . . . 5
6. Coefficient coding . . . . . . . . . . . . . . . . . . . . . 6
7. Development Repository . . . . . . . . . . . . . . . . . . . 6
8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6
9. Security Considerations . . . . . . . . . . . . . . . . . . . 6
10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 6
11. References . . . . . . . . . . . . . . . . . . . . . . . . . 7
11.1. Informative References . . . . . . . . . . . . . . . . . 7
11.2. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 7
1. Introduction
This draft describes a proposal for adapting the Opus RFC 6716
[RFC6716] energy conservation principle to video coding based on a
pyramid vector quantizer (PVQ) [Pyramid-VQ]. One potential advantage
of conserving energy of the AC coefficients in video coding is
preserving textures rather than low-passing them. Also, by
introducing a fixed-resolution PVQ-type quantizer, we automatically
gain a simple activity masking model.
The main challenge of adapting this scheme to video is that we have a
good prediction (the reference frame), so we are essentially starting
from a point that is already on the PVQ hyper-sphere, rather than at
the origin like in CELT. Other challenges are the introduction of a
quantization matrix and the fact that we want the reference (motion
predicted) data to perfectly correspond to one of the entries in our
codebook. This proposal is described in greater details in
[Perceptual-VQ], as well as in demo [PVQ-demo].
2. Gain-Shape Coding and Activity Masking
The main idea behind the proposed video coding scheme is to code
groups of DCT coefficient as a scalar gain and a unit-norm "shape"
vector. A block's AC coefficients may all be part of the same group,
or may be divided by frequency (e.g. by octave) and/or by
directionality (horizontal vs vertical).
It is desirable for a single quality parameter to control the
resolution of both the gain and the shape. Ideally, that quality
parameter should also take into account activity masking, that is,
the fact that the eye is less sensitive to regions of an image that
Valin Expires September 10, 2015 [Page 2]
Internet-Draft Video PVQ March 2015
have more details. According to Jason Garrett-Glaser, the perceptual
analysis in the x264 encoder uses a resolution proportional to the
variance of the AC coefficients raised to the power a, with a=0.173.
For gain-shape quantization, this is equivalent to using a resolution
of g^(2a), where g is the gain. We can derive a scalar quantizer
that follows this resolution:
b
g=Q_g gamma ,
where gamma is the gain quantization index, b=1/(1-2*a) and Q_g is
the gain resolution and main quality parameter.
An important aspect of the current proposal is the use of prediction.
In the case of the gain, there is usually a significant correlation
with the gain of neighboring blocks. One way to predict the gain of
a block is to compute the gain of the coefficients obtained through
intra or inter prediction. Another way is to use the encoded gain of
the neighboring blocks to explicitly predict the gain of the current
block.
3. Householder Reflection
Let vector x_d denote the (pre-normalization) DCT band to be coded in
the current block and vector r_d denote the corresponding reference
(based on intra prediction or motion compensation), the encoder
computes and encodes the "band gain" g = sqrt(x_d^T x_d). The
normalized band is computed as
x_d
x = --------- ,
|| x_d ||
with the normalized reference vector r similarly computed based on
r_d. The encoder then finds the position and sign of the largest
component in vector r:
m = argmax_i | r_i |
s = sign(r_m)
and computes the Householder reflection that reflects r to -s e_m,
where e_m is a unit vector that points in the direction of dimension
m. The reflection vector is given by
v = r + s e_m .
The encoder reflects the normalized band to find the unit-norm vector
Valin Expires September 10, 2015 [Page 3]
Internet-Draft Video PVQ March 2015
v^T x
z = x - 2 ----- v .
v^T v
The closer the current band is from the reference band, the closer z
is from -s e_m. This can be represented either as an angle, or as a
coordinate on a projected pyramid.
4. Angle-Based Encoding
Assuming no quantization, the similarity can be represented by the
angle
theta = arccos(-s z_m) .
If theta is quantized and transmitted to the decoder, then z can be
reconstructed as
z = -s cos(theta) e_m + sin(theta) z_r ,
where z_r is a unit vector based on z that excludes dimension m.
The vector z_r can be quantized using PVQ. Let y be a vector of
integers that satisfies
sum_i(|y[i]|) = K ,
with K determined in advance, then the PVQ search finds the vector y
that maximizes y^T z_r / (y^T y) . The quantized version of z_r is
y
z_rq = ------- .
|| y ||
If we assume that MSE is a good criterion for optimizing the
resolution, then the angle quantization resolution should be
(roughly)
dg 1 b
Q_theta = ---------*----- = ------ .
d(gamma) g gamma
To derive the optimal K we need to consider the normalized distortion
for a Laplace-distributed variable found experimentally to be
approximately
Valin Expires September 10, 2015 [Page 4]
Internet-Draft Video PVQ March 2015
(N-1)^2 + C*(N-1)
D_p = ----------------- ,
24*K^2
with C ~= 4.2. The distortion due to the gain is
b^2*Q_g^2*gamma^(2*b-2)
D_g = ----------------------- .
12
Since PVQ codes N-2 degrees of freedom, its distortion should also be
(N-2) times the gain distortion, which eventually leads us to the
optimal number of pulses
gamma*sin(theta) / N + C - 2 \
K = ---------------- sqrt | --------- | .
b \ 2 /
The value of K does not need to be coded because all the variables it
depends on are known to the decoder. However, because Q_theta
depends on the gain, this can lead to unacceptable loss propagation
behavior in the case where inter prediction is used for the gain.
This problem can be worked around by making the approximation
sin(theta)~=theta. With this approximation, then K depends only on
the theta quantization index, with no dependency on the gain.
Alternatively, instead of quantizing theta, we can quantize
sin(theta) which also removes the dependency on the gain. In the
general case, we quantize f(theta) and then assume that
sin(theta)~=f(theta). A possible choice of f(theta) is a quadratic
function of the form:
2
f(theta) = a1 theta - a2 theta.
where a1 and a2 are two constants satisfying the constraint that
f(pi/2)=pi/2. The value of f(theta) can also be predicted, but in
case where we care about error propagation, it should only be
predicted from information coded in the current frame.
5. Bi-prediction
We can use this scheme for bi-prediction by introducing a second
theta parameter. For the case of two (normalized) reference frames
r1 and r2, we introduce s1=(r1+r2)/2 and s2=(r1-r2)/2. We start by
using s1 as a reference, apply the Householder reflection to both x
and s2, and evaluate theta1. From there, we derive a second
Householder reflection from the reflected version of s2 and apply it
to z. The result is that the theta2 parameter controls how the
Valin Expires September 10, 2015 [Page 5]
Internet-Draft Video PVQ March 2015
current image compares to the two reference images. It should even
be possible to use this in the case of fades, using two references
that are before the frame being encoded.
6. Coefficient coding
Encoding coefficients quantized with PVQ differs from encoding
scalar-quantized coefficients from the fact that the sum of the
coefficients magnitude is known (equal to K). It is possible to take
advantage of the known K value either through modeling the
distribution of coefficient magnitude or by modeling the zero runs.
In the case of magnitude modeling, the expectation of the magnitude
of coefficient n is modeled as
K_n
E(|y_n|) = alpha * ----- ,
N - n
where K_n is the number of pulses left after encoding coeffients from
0 to n-1 and alpha depends on the distribution of the coefficients.
For run-length modeling, the expectation of the position of the next
non-zero coefficient is given by
N - n
E(|run|) = beta * ----- ,
K_n
where beta also models the coefficient distribution.
7. Development Repository
The algorithms in this proposal are being developed as part of
Xiph.Org's Daala project. The code is available in the Daala git
repository at [1]. See [2] for more information.
8. IANA Considerations
This document makes no request of IANA.
9. Security Considerations
This draft has no security considerations.
10. Acknowledgements
Thanks to Jason Garrett-Glaser, Timothy Terriberry, Greg Maxwell, and
Nathan Egge for their contribution to this document.
Valin Expires September 10, 2015 [Page 6]
Internet-Draft Video PVQ March 2015
11. References
11.1. Informative References
[PVQ-demo]
Valin, JM., "Daala: Perceptual Vector Quantization (PVQ)",
November 2014, <https://people.xiph.org/~jm/daala/
pvq_demo/>.
[Perceptual-VQ]
Valin, JM. and TB. Terriberry, "Perceptual Vector
Quantization for Video Coding", Proceedings of SPIE Visual
Information Processing and Communication , February 2015,
<http://jmvalin.ca/papers/spie_pvq.pdf>.
[Pyramid-VQ]
Fischer, T., "A Pyramid Vector Quantizer", IEEE Trans. on
Information Theory, Vol. 32 pp. 568-583, July 1986.
[RFC6716] Valin, JM., Vos, K., and T. Terriberry, "Definition of the
Opus Audio Codec", RFC 6716, September 2012.
11.2. URIs
[1] https://git.xiph.org/daala.git
[2] https://xiph.org/daala/
Author's Address
Jean-Marc Valin
Mozilla
331 E. Evelyn Avenue
Mountain View, CA 94041
USA
Email: jmvalin@jmvalin.ca
Valin Expires September 10, 2015 [Page 7]
