Idoia Ochoa and Nima Soltani Outline System overview Detailed encoder description Demonstration Results Extensions Conclusions System Overview Encoder R L DWT Quant Arith ID: 728431
Download Presentation The PPT/PDF document "Encoding Stereo Images Christopher Li," is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Encoding Stereo Images
Christopher Li,
Idoia
Ochoa
and
Nima
Soltani
Slide2
Outline
System overview
Detailed encoder description
Demonstration
Results
Extensions
ConclusionsSlide3
System Overview (Encoder)
R
L
DWT
Quant
Arith
Enc
DWT
Quant
Motion Estimation
DCT
Re-order
Arith
Enc
Arith
Enc
Huff
Enc
residuals
shift vectors
u
se MESlide4
Left Image
Daubechies-4 wavelet decomposition
5 levels for luminance, 4 for chrominance
Uniform quantization with adaptive levels
Each component meets its own fraction of MSE
Arithmetic coding on the quantized residuals
Frequency tables are sent for each arithmetic coderSlide5
Left Quantization
Decomposed PSNR constraint
Allocated fractions of MSE to each color component
Met PSNR constraints by finding maximum uniform quantization levels that meet assigned MSEs
Slide6
Left Quantization
Motion Estimation Enable Signal
Heuristically choose differential vs.
separate encoding of right image
Quantize with
Calculate MSE
Y
wavelet
coeffs
Encode differentially
Encode separately
Yes
NoSlide7
Right Image
Motion Estimation Block
Partition into
30x30
blocks
Find shift vectors that minimize the
MSESearch an area from [-64,64] in the direction and [-6,6] in the direction for minimum distortion
Slide8
Right Image
Residual coding
Impose residuals of
Cb
and Cr to be 0
Use remaining fraction of MSE for Y component
Compute DCT of blockReshape using zig-zag orderingReplace remaining zeros in block with end of block characterPerform arithmetic codingSlide9
Right Image
Shift vector coding
Offline
Find joint statistics of the shift vectors over the training set
Construct Huffman
table
During run-time, encode shift vectors using this Huffman tableSlide10
Right Image
Separately coded
Same method as left image
D4 wavelet, with 5 levels for Y, 4 for
Cb
, Cr
Uniform quantization with variable stepArithmetic coding with frequencies sentSlide11
Writing to File
Unique quantization values encoded in header bits
Arithmetic coders
Encode frequencies, output
length of
sequence and sequence itself
Huffman encoders Length of sequence and sequence itselfTables stored offlineSlide12
Decoder
Perform all the steps of the encoder in reverse
Decode left image using inverse DWT
Read
motion estimation
flag for right image
If enabled, decode shift vectors and residualsElse, decode using inverse DWTSlide13
DemonstrationSlide14
Results
Image
Bits/pixel
1
1.7324
2
0.7566
30.1807
4
0.716450.930461.239071.8104
Image
Bits/pixel
81.71889
0.6453101.776611
0.9341120.587913
2.1404142.3837Slide15
Block sizeSlide16
Extensions
Use intra-block coding for right image
Explore using
d
ifferent wavelets
Implement embedded zero trees in C
Explore run-length coding furtherApply uniform deadzone quantizersSlide17
Conclusions
Important trade-off between bits allocated to shift data and residual data
Arithmetic coding outperforms Huffman
Reshaping the DCT blocks allows us to use information, such as its size, to our advantage
Uniform
quantizer
is faster, simpler and has less overhead than Lloyd-max quantizersMEX files reduce runtime significantly!Slide18
Thank you
Questions?