Ankur Saxena

About Me

I am at Nvidia, California. Earlier, I was with Samsung Research America-Dallas, Texas. Even earlier, I completed my PhD from the ECE department at the University of California-Santa Barbara (UCSB). My PhD advisor was Prof Kenneth Rose and I was affiliated with the Signal Compression Laboratorywhile at UCSB.

I did my B.Tech from Indian Institute of Technology (IIT) -Delhi in 2003. During my PhD, I interned at NTT Docomo Labs, Palo Alto while I interned at Fraunhofer Labs, Erlangen, Germany in my undergraduate days.

Education

• Ph.D., University of California-Santa Barbara, Electrical and Computer Engineering, 2005-2008.

• M.S., University of California-Santa Barbara, Electrical and Computer Engineering, 2003-2004.

• B.Tech., Indian Institute of Technology-Delhi, Electrical Engineering, India, 1999-2003.

Work Highlights

• 9 years industry and standards experience in the fields of video coding and processing, image search, HDR imaging, virtual reality, video fingerprinting, with 14 years of overall research experience.

• Multiple standard contributions adopted in HEVC video coding standard, and its extensions; China AVS 2.0 video standard, compact descriptors for visual search (CDVS) in MPEG.

• Coordinator for 7 Core Experiments in HEVC and its extensions; and CDVS Standards in MPEG/JCTVC.

• Multiple invented techniques in video coding, virtual reality, and video fingerprinting have been part of company IP's products such as MilkVR service, and TV product lines.

• Created several internal company funding proposals with Sr. Directors/VP’s in company.

• Inventor/Co-Inventor of about 15 patent applications.

• Authored or co-authored about 40 academic papers; and 45 standards contributions.

• Mentored multiple interns over years; and managed university collaboration projects.

• Recipient of multiple best paper awards, and numerous internal company awards:

  • IEEE Signal Processing Society Young Author Best Paper Award, 2015.

  • Top 10% paper award at ICIP 2014.

  • Samsung Prolific Author Award, 2013.

  • Samsung Best Paper Award, 2011.

  • Best Student Paper Finalist, ICASSP 2009

Research Summary

My research interests are communication, signal and processing and computer vision. I have worked in the areas of video and image compression, distributed source coding, signal processing for sensor networks, video fingerprinting, and image search.

I have represented Samsung as an official delegate at ITU-T and ISO MPEG: HEVC video standardization and CDVS meetings, and have worked extensively on various algorithms for HEVC and CDVS standards.

My work on transform coding: on DCT/DST technology was adopted in HEVC Standardization.The proposed DST Type-7 is the only transform other than the conventional DCT in the HEVC video coding standard.

I also received the Samsung Best Paper Award for the year 2011. The competition had 1043 submissions in different fields from various Samsung Offices throughout the world. I also received Prolific Author of the year award for the year 2013 at Samsung Research America, Dallas.

I am also a co-inventor of secondary transforms which can be applied after DCT for both intra and inter residues. The secondary transforms is part of the China AVS 2.0 video coding standard. I have also worked on video fingerprinting, and am currently exploring various aspects of visual search in the CDVS standardization project.

At UCSB, my PhD thesis was on 'Distributed coding of spatio-temporally correlated sources' . It focuses on solving various major problems in distributed coding (More details are available here).

For part of my PhD work in ICASSP 2009, I was a Finalist in the Student Best Paper contest at ICASSP 2009. I was a co-author of a top 10% paper at IEEE ICIP 2014, and have recently been awarded the IEEE Signal Processing Society Young Author award in 2015.

Professional Activities

Coordinator for 7 Core Experiments in MPEG/JCTVC: On Transform Skipping; intra transform mode dependency simplifications (HEVC); intra prediction improvements, combined inter and inter-layer prediction (S-HEVC); intra prediction techniques, palette coding for screen content (HEVC-Range Extensions); feature point selection of the compact descriptors for visual search standardization in MPEG.

Reviewer for IEEE Trans. on Circuits, Systems and Vehicular Technology, Image Processing, Multimedia, Communication, Wireless Communications; Springer Signal, Image and Video Processing Journal; Eurasip Journal on Advances in Signal Processing, and numerous other conferences: IEEE ICIP, ICASSP, PCS, MMSP, ISIT, IPSN.

Member of IEEE PCS 2018 Best Paper Award Selection Committee.

Membership: IEEE Senior Member, IEEE Signal Processing Society, MPEG, JCTVC.

Publications (sorted by area)

Video/Image Compression

1. Low-complexity separable multiplierless loop filter for video coding, A. Saxena, M. Aabed, and M. Budagavi, IEEE ICIP, Oct 2015.

2. Coding efficiency improvements beyond HEVC with known tools, A. Alshin; E. Alshina; M. Budagavi; K. Choi; J. Min; M. Mishourovsky; Y. Piao; A. Saxena, SPIE Applications of Digital Image Processing, Aug 2015.

3. Improvements on Intra Block Copy in natural content video coding, H. Chen, Y-S. Chen, M-T. Sun, A. Saxena, M. Budagavi, IEEE ISCAS, May 2015.

4. Nearest-neighbor intra prediction for screen content video coding, H. Chen, A. Saxena and F. Fernandes, IEEE ICIP Oct 2014. (Top 10% Paper Recognition at ICIP.)

5. On prediction techniques for palette coding, G. Jin, A. Saxena, and F. Fernandes, IEEE ICIP Oct 2014.

6. Fast secondary transforms for scalable video coding, A. Saxena and F. Fernandes, IEEE PCS, Dec 2013.

7. On secondary transforms for scalable video coding, A. Saxena and F. Fernandes, IEEE VCIP, Nov 2013.

8. DCT/DST- based transform coding for intra prediction in image/video coding, A. Saxena and F. Fernandes, IEEE Trans. on Image Processing, Oct 2013.

9. Low-latency secondary transforms for intra/inter prediction residual, A. Saxena and F. Fernandes, IEEE Trans. on Image Processing, Oct 2013.

10. Fast transforms for intra-prediction based image and video coding, A. Saxena, F. Fernandes and Y. Reznik, IEEE DCC, March 2013.

11. On secondary transforms for prediction residual, A. Saxena and F. Fernandes, IEEE ICIP, Oct 2012.

12. Jointly optimized spatial prediction and block transform for video and image coding, J. Han, A. Saxena, V. Melkote and K. Rose, IEEE Transactions on Image Processing, April 2012 (IEEE Signal Processing Society Young Author Best Paper Award)

13. On secondary transforms for intra prediction residual, A. Saxena and F. Fernandes, IEEE ICASSP, March 2012.

14. Mode Dependent DCT/DST for Intra Prediction in Block-Based Image/Video Coding, A. Saxena and F. Fernandes, IEEE ICIP, Sept 2011.

15. Towards jointly optimal spatial prediction and adaptive transform in video/image coding, J. Han, A. Saxena, and K. Rose, IEEE ICASSP, March 2010.

16. On optimal royalty costs for video compression, A. Saxena, O. Guleryuz and R. Civanlar, IEEE ICIP, Oct 2008.

17. Improving intra prediction in high efficiency video coding, H. Chen, T. Zhang, M-T. Sun, A. Saxena, M. Budagavi (Under Review)

Virtual Reality, Image Search and Video Fingerprinting

18. Motion estimation and compensation for fisheye warped video, G. Jin, A. Saxena, and M. Budagavi, IEEE ICIP, Oct 2015.

19. 360 degrees video coding using region adaptive smoothing, M. Budagavi, J. Furton, G. Jin, A. Saxena, J. Wilkinson, A. Dickerson, IEEE ICIP 2015.

20. Mid-Level Feature Based Local Descriptor Selection for Image Search, S. Bucak, A. Saxena, A. Nagar, F. Fernandes, K-P. Bhat, IEEE VCIP workshop, Nov 2013.

21. Low complexity image matching using color based SIFT, A. Nagar, A. Saxena, S. Bucak, F. Fernandes, K.-P Bhat, IEEE VCIP workshop, Nov 2013

22. Perceptual similarity based robust low-complexity video fingerprinting, K. Vadivel, F. Fernandes, Z. Ma, P. Lai and A. Saxena, IEEE ICASSP, March 2012.

Source Coding and Sensor Networks

23. Error/Erasure-Resilient and Complexity-Constrained Zero-Delay Distributed Coding for Large Scale Sensor Networks, K. Viswanatha, S. Ramaswamy, A. Saxena and K. Rose, ACM Trans. on Sensor Networks (TOSN), Feb 2015.

24. Error-Resilient and Complexity-Constrained Distributed Coding for Large Scale Sensor Networks, K. Viswanatha, S. Ramaswamy, A. Saxena and K. Rose, ACM/ IEEE IPSN, April 2012 (22 papers accepted out of 147 submissions).

25. A classifier based decoding approach for large scale distributed coding, K. Viswanatha, S. Ramaswamy, A. Saxena and K. Rose, IEEE ICASSP, May 2011.

26. On scalable distributed coding of correlated sources, A. Saxena and K. Rose, IEEE Trans. on Signal Processing, May 2010.

27. Towards large scale distributed coding, S. Ramaswamy, K. Viswanatha, A. Saxena and K. Rose, IEEE ICASSP, March 2010.

28. Robust distributed source coder design by deterministic annealing, A. Saxena, J. Nayak and K. Rose, IEEE Trans. on Signal Processing, Feb 2010.

29. Towards large scale distributed coding of correlated sources, S. Ramaswamy, K. Viswanatha, A. Saxena and K. Rose, Southern California Workshop on Distributed Multimedia Systems, Nov 2009.

30. On distributed and scalable quantization of correlated sources, A. Saxena and K. Rose, Southern California Workshop on Distributed Multimedia Systems, Nov 2009.

31. Scalable distributed source coding, A. Saxena and K. Rose, IEEE ICASSP, April 2009 (Best Student Paper Finalist, Top 1%)

32. Distributed predictive coding for spatio-temporally correlated sources, A. Saxena and K. Rose, IEEE Trans. on Signal Processing, Oct 2009.

33. Optimization of correlated source coding for event-based compression in sensor networks, J. Singh, A. Saxena, K. Rose and U. Madhow, IEEE DCC, March 2009.

34. On distributed quantization in scalable and predictive coding, A. Saxena and K. Rose, Proc. Sensor, Signal and Information Processing (SENSIP), May 2008.

35. Distributed multi-stage coding of correlated sources, A. Saxena and K. Rose, IEEE DCC, March 2008.

36. Challenges and recent advances in distributed predictive coding, A. Saxena and K. Rose (Invited Paper), IEEE ITW, Sept 2007.

37. Distributed predictive coding for spatio-temporally correlated sources, A. Saxena and K. Rose, IEEE ISIT, June 2007.

38. A global approach to joint quantizer design for distributed coding of correlated sources, A. Saxena, J. Nayak and K. Rose, IEEE ICASSP, May 2006.

39. On efficient quantizer design for robust distributed source coding, A. Saxena, J. Nayak and K. Rose, IEEE DCC, March 2006.

40. On large scale distributed compression and communication: Part I: Design of low-complexity decoders, K. Viswanatha, S. Ramaswamy, A. Saxena, E. Akyol and K. Rose (Under Review)

41. On large scale distributed compression and communication: Part II: Dispersive information routing, K. Viswanatha, S. Ramaswamy, A. Saxena, E. Akyol and K. Rose (Under Review)

Patents (sorted by area)

Video/Image Compression

1. Methods for in-loop filtering in video coding, A. Saxena, M. Aabed, M. Budagavi, 2015

2. Method for intra prediction improvements for oblique modes in video coding, A. Saxena, H. Chen, F. Fernandes, 2014.

3. Method and apparatus for applying secondary transforms on enhancement-layer residuals, A. Saxena and F. Fernandes, 2014.

4. Methods for palette prediction and intra block copy padding, A. Saxena, G. Jin, F. Fernandes, 2014

5. Mode-dependent transforms for residual coding with low latency, US 20130003856 A1, A. Saxena and F. Fernandes, 2012

6. Low complexity transform coding using adaptive DCT/DST for intra-prediction, A. Saxena and F. Fernandes, 2011, US patent granted.

Virtual Reality and High Dynamic Range

7. Methods for generating and transmitting metadata for virtual reality, A. Saxena, H.N.-Zadeh, M. Budagavi, 2015

8. Methods and apparatus for enhancing images via white pop-out, A. Saxena, H.N.-Zadeh, M. Budagavi, 2015

9. Coding of 360 degrees videos using region adaptive smoothing, J. Furton, G. Jin, J. Wilkinson, A. Dickerson, M. Budagavi, A. Saxena, 2015

10. View-Dependent Tone Mapping for Virtual Reality, H. N.-Zadeh, M. Budagavi, A. Saxena, 2015.

11. Video Enhancement via Light Source Compensation, H. N.-Zadeh, M. Budagavi, A. Saxena, 2015

Image Search and Fingerprinting

12. Novel criteria for Gaussian mixture model cluster selection in scalable compressed fisher vector (scfv) global descriptor, G. Srivastava, Z. Li, A. Nagar, A. Saxena, Z. Ma and F. Fernandes, 2014

13. Apparatus and method for performing visual search, A. Saxena, S. Bucak, A. Nagar, F. Fernandes, G. Srivastava, 2013

14. Incremental visual query processing with holistic feature feedback, Z. Li, A. Saxena, A. Nagar, G. Srivastava, K. P. Bhat, 2013

15. Apparatus and method for robust low-complexity video fingerprinting, F. Fernandes, K. Vadivel, Z. Ma, P. Lai, and A. Saxena, 2012, US patent granted.

Standard Contributions

(All the following contributions are for Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T SG16 WP3 and ISO/IEC JTC1/SC29/WG11 for the HEVC video coding standardization)

38. Palette prediction for palette coding, G. Jin, A. Saxena, F. Fernandes, San Jose, USA, Jan 2014.

37. Simplification of major color based palette prediction, G. Jin, A. Saxena, F. Fernandes, San Jose, USA, Jan 2014.

36. Summary report of HEVC Range Extensions Core Experiments 4 (RCE4) on palette coding for screen content, X. Guo, A. Saxena, San Jose, USA, Jan 2014

35. On intra block copy motion vector coding, G. Jin, A. Saxena, F. Fernandes, Geneva, Switzerland, October 2013.

34. On transform selection for Intra-BlockCopy blocks A. Saxena, E. Alshina, F. Fernandes, Geneva, Switzerland, October 2013.

33. Combination of sample adaptive prediction and nearest neighbor prediction for oblique modes, A. Saxena, H. Chen, F. Fernandes, Geneva, Switzerland, October 2013.

32. Nearest-neighbor intra prediction for screen content video coding, H. Chen, A. Saxena, F. Fernandes, Geneva, Switzerland, October 2013.

31. HEVC Range Extensions Core Experiment 4 (RCE 4): Palette Coding For Screen Content L. Guo, X. Guo, A. Saxena, Geneva, Switzerland, October 2013.

30. Summary report of HEVC Range Extensions Core Experiment 3 on Intra Prediction techniques, A. Saxena, D. Kwon, M. Naccari, C. Pang, Geneva, Switzerland, October 2013.

29. On sample adaptive intra prediction for oblique modes in lossless coding, H. Chen, A. Saxena and F. Fernandes, Vienna, Austria, July 2013.

28. On sample adaptive intra prediction for oblique modes in lossy coding, A. Saxena, H. Chen and F. Fernandes, Vienna, Austria, July 2013.

27. Enhanced angular intra prediction for screen content coding, H. Chen, A. Saxena and F. Fernandes, Vienna, Austria, July 2013.

26. HEVC Range Extensions Core Experiment 3 (RCE3): Intra Prediction techniques, A. Saxena, D. Kwon, M. Naccari, C. Pang, Vienna, Austria, July 2013.

25. On secondary transforms for Intra_BL residue, A. Saxena and F. Fernandes, Incheon, Korea, April 2013.

24. On estimation theoretic prediction for enhancement layer residual in scalable video coding, A. Saxena and F. Fernandes, Geneva, Switzerland, Jan 2013

23. Summary Report of Tool Experiment on Combined Prediction in SHVC, X. Li, E. François, P. Lai, D.-K. Kwon, A. Saxena, Geneva, Switzerland, Jan 2013

22. Description of Core Experiment 1: Intra Prediction Improvements in SHVC, A. Tabatabai, K. Rapaka, A. Saxena, S. Liu, Geneva, Switzerland, Jan 2013

21. Description of Core Experiment 3: Combined Inter and Inter-Layer Prediction in SHVC, X. Li, E. Francois, P. Lai, D. Kwon, A. Saxena, Geneva, Switzerland, Jan 2013.

20. Description of Tool Experiment B3: Combined Prediction in SHVC, X. Li, E. Francois, P. Lai, D. Kwon, A. Saxena, Shanghai, China, Oct 2012.

19. Summary report of core experiment on intra transform mode dependency simplifications, K. Ugur and A. Saxena, Stockholm, Sweden, July 2012.

18. Break out Group: CE1 visual test report, K. Ugur and A. Saxena, Stockholm, Sweden, July 2012.

17. Description of Core Experiment 1 (CE1): Intra transform mode dependency simplifications, K. Ugur and A. Saxena, Geneva, Switzerland, May 2012.

16. On secondary transforms for intra/inter prediction residual, A. Saxena and F. Fernandes, Geneva, Switzerland, May 2012.

15. Recent results for secondary transforms for intra/inter prediction residual, A. Saxena, Y. Shibahara, E. Alshina, F. Fernandes and T. Nishi, San Jose, USA, Feb 2012.

14. Summary Report of Core Experiment on Transform Skipping, M. Mrak, J. Sole, J. Xu and A. Saxena, San Jose, USA, Feb 2012.

13. On secondary transforms for intra prediction residual, A. Saxena, Y. Shibahara, F. Fernandes and T. Nishi, San Jose, USA, Feb 2012.

12. On secondary transforms for inter prediction residual, A. Saxena, Y. Shibahara, F. Fernandes and T. Nishi, San Jose, USA, Feb 2012.

11. Harmonization of SDIP and mode-dependent secondary transforms, A. Saxena, Y. Shibahara, F. Fernandes and T. Nishi, Geneva, Switzerland, Nov 2011.

10. On secondary transforms for inter prediction residual, A. Saxena and F. Fernandes, Geneva, Switzerland, Nov 2011.

9. Description of core experiment 5 (CE5): Transform Skipping, M. Mrak, J. Sole, J. Xu, A. Saxena, Geneva, Switzerland, Nov 2011.

8. On secondary transforms for intra prediction residual, A. Saxena and F. Fernandes, Torino, Italy, July 2011.

7. Mode-Dependent DCT/DST for 4x4 Chroma Blocks, A. Saxena, F. Fernandes, E. Alshina and J. Chen, Torino, Italy, July 2011.

6. On fast implementation of 4-point DST Type-7 with 5 multiplications, A. Saxena and F. Fernandes, Torino, Italy, July 2011.

5. Mode-Dependent 8x8 DCT/DST for Intra Prediction, A. Saxena and F. Fernandes, Torino, Italy, July 2011.

4. Mode-dependent DCT/DST without 4*4 full matrix multiplication for intra prediction, A. Saxena and F. Fernandes, Geneva, Switzerland, March 2011.

3. Experimental results of Rotational Transforms (ROT), E. Alshina, A. Alshin, F. Fernandes, A. Saxena, V. Seregin, Z. Ma, W. J. Han, Geneva, Switzerland, March 2011.

2. Mode-dependent DCT/DST for intra prediction in video coding, A. Saxena and F. Fernandes, Daegu, Korea, Jan 2011.

1. Jointly optimal intra prediction and adaptive primary transform, A. Saxena and F. Fernandes, Guangzhou, China, Oct 2010.

(The following contributions are for MPEG Compact Descriptors for Visual Search (CDVS) standardization)

6. SKIP Mode - Reconstructing Global Descriptor from Local Descriptors at Server End, A. Nagar, X. Xin, Z. Ma, Z. Li, G. Srivastava, A. Saxena, Y. Lim, F. Fernandes, K. Bhat, MPEG CDVS, Geneva, Switzerland, Jan 2013.

5. Improvements to global descriptors using higher order moments, G. Srivastava, A. Saxena, A. Nagar, S. Bucak and F. Fernandes, MPEG CDVS, Shanghai, China, October 2012.

4. Improved method for image search from local descriptors with visual meaning score, S. Bucak, A. Saxena, A. Nagar, G. Srivastava, F. Fernandes, K. Bhat, MPEG CDVS, Shanghai, China, October 2012.

3. Incremental Query Processing with a Holistic Feature Feedbacks, Z.Li, A. Saxena, A. Nagar, G. Srivastava, K. Bhat, MPEG CDVS, Shanghai, China, October 2012.

2. Using color information in local descriptors, A. Nagar, A. Saxena, S. Bucak, F. Fernandes, K. Bhat, MPEG CDVS, Stockholm, Sweden, July 2012.

1. Descriptor Extraction and Matching using Visual Attention Region for CDVS, L. Cheok, J. Song, K. Park and A. Saxena, MPEG CDVS, Geneva, Switzerland, May 2012.

(Other standard contributions)

2. Use Cases and Requirements for Omnidirectional Media Format, B. Choi, G. Lee, K. Park, A. Saxena, M. Budagavi, Y. Lim, Geneva, Switzerland, MPEG-Requirements, Oct 2015.

1. Known tools performance investigation for next generation video coding, E. Alshina, A. Alshin, J. Min, K. Choi, A. Saxena, M. Budagavi, VCEG-AZ05, Warsaw, Poland, June 2015.