Derek Pang
Email: dcypang@alumni.stanford.edu
Updated: February 2024.
About Me
Derek Pang is currently a Tech Lead Manager at Google, working on Pixel HDR+ with focus on computational photography and on-device machine learning. He helped the team ship various photo processing features, including Night Sight, 50MP high-resolution mode, UltraHDR, VideoBoost, Panorama, and ML denoise/deblur.
Prior to Pixel Camera, he led a collaboration with Google DeepMind on learning-to-encode as part of the YouTube Media Algorithm team. Prior to joining Google, he was Video Architect at Lytro, where he worked on numerous inventions in video compression and streaming to enable cinematic light field VR content delivery. From 2012 to 2015, he was a founding member of Highfive Technologies and designed one of the first WebRTC cloud-based video conferencing systems for the enterprise market.
Derek earned his M.S. in Electrical Engineering from Stanford University. His research interests included computational photography, on-device ML, multiview and video compression. During his PhD candidacy at Stanford, he co-developed ClassX, an open source interactive video streaming system aimed at reducing the cost of capturing and publishing educational videos.
[Resume] [Google Scholar] [Linkedin]
Projects
Pixel 7 Pro to 9 Pro, Foldable, A-series
Launched numerous HDR+ processing innovations that:
Enhanced lowlight photography: 50MP NightSight / Astrophotography
Expanded creative control: New Panorama, VideoBoost, Pro control, UltraHDR,
Unparalleled imaging quality: extended dynamic range via exposure fusion, ML denoise / deblur.
Our team effort enabled Pixel devices to be best-in-class in photo IQ in their respective market segments:
#1 DxO Photo Score in US market: Pixel 7 Pro, 8 Pro, and 9 Pro .
#1 DxO Photo Score in foldable market: Pixel 9 Pro Fold, Pixel Fold
#1 DxO Photo Score in <$400 Segment: Pixel 7a, 8a, 9a.
Google Pixel 3 & Pixel 4
The winner of Innovation of the Year Award in 2018 and 2019 from DPReview. My contribution on Pixel 3 included:
Launched the first HEVC support in Google Camera to reduce video storage
Optimized codec quality to achieve best-in-class video capture experience.
Led cross-functional team in building automated and large-scale video quality validation infrastructure.
Designed a novel power-efficient entropy-adaptive pre-filtering technique to mitigate compression artifacts for low bit-rate applications.
Worked with vendors and partners to advance video compression quality across Pixel and Android ecosystem.
Developed algorithms for multi-camera parameter estimation and view synthesis.
Lytro Immerge / 6-DoF Video Codec
World's First Volumetric Cinematic VR system using light field technology. Developed a proprietary 6-DoF video codec that reduces data rate by more than 250x, while maintaining real-time 90fps playback performance. Our Lytro Immerge system has won: "Technology Awards for outstanding achievement in Entertainment Arts and Technologies" from Advanced Imaging Society in 2018.
[Wired] [VR Capture Format] [Hallelujah in VR] [SIGGRAPH] [Patents]
Highfive Cloud-Based Video Conferencing
An easy-to-use, affordable conferencing solution that has helped more than 1,400 businesses and served over one million call minutes per week in its first year of launch. Technical work included proprietary rate control, multiparty simulcast, burst lost resilient forward error correction, error concealment, jitter buffer and cross-platform media pipeline.
[Wired] [Bloomberg] [The Verge] [MarketWired] [Redot Design Award Winner 2015]
ClassX Open Source Project
An open-source interactive video streaming project in online education. A precursor technology to tiled 360/VR streaming.
[Source] [Paper] [Stanford News, Stanford Daily, ACM TechNews]
Publications
H Maia, J Qiu, WS Lai, W Chow, S Kuroda, J Klopp, A Kae, R Geiss, F Bleibel, S Tsai, D Pang, Y Shih Highlight Recovery Using Multi-exposure Fusion and Floating-point Tone-mapping. TD Common. 2024.
H. Mao, C. Gu, M. Wang, A. Chen, N. Lazic, N. Levine, D. Pang, R. Claus, M. Hechtman, C. Chiang, C. Chen, J. Han. Neural Rate Control for Video Encoding using Imitation Learning. ICML workshop 2021. [Poster]
Mandhane, A., Zhernov, A., Rauh, M., Gu, C., Wang, M., Xue, F., Shang, W., Pang, D., Claus, R., Chiang, C.H. and Chen, C., 2022. "Muzero with self-competition for rate control in vp9 video compression". arXiv preprint arXiv:2202.06626.
D. Pang , S. Halawa, N.-M. Cheung and B. Girod, "Mobile Interactive Region-of-Interest Video Streaming with Crowd-Driven Prefetching," International ACM Workshop on Interactive Multimedia on Mobile and Portable Devices, ACM Multimedia (MM’11), Scottsdale, Arizona, USA, November 2011.
D. Pang , S. Halawa, N.-M. Cheung and B. Girod, "ClassX Mobile - Region-of-Interest Video Streaming to Mobile Device with Multi-Touch Interaction," ACM Multimedia (MM’11), Scottsdale, Arizona, USA, November 2011. [Demo]
S. Halawa, D. Pang, N.-M. Cheung and B. Girod, "ClassX – An Open Source Interactive Lecture Streaming System," ACM Multimedia (MM’11), Scottsdale, Arizona, USA, November 2011. (Open Source Software Competition Honorable Mention). [source code]
M. Makar, YC Lin, N.-M. Cheung, D. Pang and B. Girod, "Quality-Controlled View Interpolation for Multiview Video," IEEE International Conference on Image Processing (ICIP 2011), Brussels, Belgium, September 2011.
N.-M. Cheung, S. Halawa, D. Pang and B. Girod, "Multimedia Technology for Next Generation Online Lecture Video," IEEE Communications Society Multimedia Communications Technical Committee (MMTC), e-Letter, Vol.6 , No.1, January 2011.
X. Xiu, D. Pang and J. Liang, "Projective Rectification-based View Interpolation and Extrapolation for Multiview Video Coding and Free Viewpoint Video," IEEE Transactions on Circuits and Systems for Video Technology, Vol. 21, No. 6, June 2011.
A. Mavlankar, P. Agrawal, D. Pang, S. Halawa, N.-M. Cheung, and B. Girod, "An Interactive Region-of-Interest Video Streaming System for Online Lecture Viewing," Special Session on Advanced Interactive Multimedia Streaming, Proc. of 18th International Packet Video Workshop (PV), Hong Kong, China, December 2010. (Invited paper).
M. Makar, D. Pang , Y.C. Lin , and B. Girod, "Quality-controlled Motion-compensated Interpolation," Proc. 44th Annual Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, USA, November 2010. (Invited paper).
D. Pang, X. Xiu, and J. Liang, "Multiview Video Coding Using Projective Rectification-Based View Extrapolation and Synthesis Bias Correction," Proc. IEEE International Conference on Multimedia and Expo, ICME 2009, New York, USA, June 2009. (Student Travel Award).
A. Kimura, D. Pang, T. Takeuchi, J. Yamato and K. Kashino, "Dynamic Markov random field for stochastic modeling of visual attention," Proc. International Conference on Pattern Recognition, ICPR2008, Tempa, Flordia, USA, December 2008.
D. Pang, A. Kimura, T. Takeuchi, J. Yamato and K. Kashino, "A stochastic model of selective visual attention with a dynamic Bayesian network," Meeting on Image Recognition and Understanding, MIRU2008, Karuizawa, Nagano, Japan, July 2008. (Best Interactive Session Award).
D. Pang, A. Kimura, T. Takeuchi, J. Yamato and K. Kashino, "A stochastic model of selective visual attention with a dynamic Bayesian network," Proc. IEEE International Conference on Multimedia and Expo, ICME 2008, Hannover, Germany, June 2008.
Patents
Rate control machine learning models with feedback control for video encoding. C Gu, H Mao, C Chiang, C Chen, J Han, D Pang, R Claus, M Hechtman, D James Visentin, C Fougner, CSchaff, N Patil, A Bellido. US Patent 12,088,823, 2024.
Methods and systems for encoder parameter setting optimization. CYD Pang, K Felder, A Gadde, P Wilkins, C Chen, YC Lin. US Patent App. 18/403,659.
Methods and systems for encoder parameter setting optimization. CYD Pang, K Felder, A Gadde, P Wilkins, C Chen, YC Lin. US Patent 11,870,833. 2024
Adaptive GOP structure using temporal dependencies likelihood. CYD Pang, Y Su, YC Lin - US Patent 11,363,262, 2022
Adaptive control for immersive experience delivery. A Song, D Pang, M Ma, N Karnad - US Patent App. 15/590,808, 2018.
Video compression with adaptive view-dependent lighting removal. D Pang, C Pitts, K Akeley - US Patent App. 15/832,023, 2018.
Layered content delivery for virtual and augmented reality experiences. D Pang, C Pitts, K Akeley - US Patent App. 15/729,918, 2018.
Encoding and decoding virtual reality video. D Pang, C Pitts, K Akeley, Z Htet - US Patent App. 15/730,096, 2018.
Data structures and delivery methods for expediting virtual reality playback. D Pang, C Pitts, K Akeley - US Patent App. 15/706,343, 2018.
Spatial random access enabled video system with a three-dimensional viewing volume. D Pang, C Pitts, K Akeley - US Patent App. 15/590,877, 2017.
Proximity-based conference session transfer. Shantanu Sinha, ROY Jeremy, Ching Yin Derek Pang, Ohene Kwasi Ohene-adu, Edward Wei, Sayli Anand Benadikar, Sankara Narayana Hemanth Meenakshisundaram- US Patent App. 14/726,271, 2015.
Method and system for multiparty video conferencing. ROY Jeremy, CYD Pang, OK Ohene-adu, E Wei… - US Patent App. 14/726,307, 2015.