Sintel

Levy, Colin; Roosendaal, Ton

doi:10.1145/1900264.1900346

Cited by 5 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We benchmarked FLAC using Ghosts I-IV (Nine Inch Nails, 2008), a music album of a band 'Nine Inch Nails', that contains 36 tracks of improvisation music, released under the Creative Commons license. We benchmarked x264 using a trailer of Sintel (Levy, C., 2010), an open-content film created and released under the Creative Commons license by Blender Foundation.…”

Section: Subject Software Systemsmentioning

confidence: 99%

Transferring Pareto Frontiers across Heterogeneous Hardware Environments

Valov

Guo

Czarnecki

2020

Proceedings of the ACM/SPEC International Conference on Performance Engineering

View full text Add to dashboard Cite

Further, I would like to thank my research collaborators Jean-Christophe Petkovich and Dr. Jianmei Guo for their engineering and scientific knowledge, suggestions, and help in this project.I would also like to thank Professor Derek Rayside, Professor Grant Weddell, Professor Jörg Andreas Kienzle, and Professor Peter van Beek for being my examining committee, for their reviews, feedback, and questions that helped me to expand and improve my work.I would like to thank our Graduate Coordinator Paula Roser, for her patience, help, and constant willingness to help me in any situation.I would like to express my sincere gratitude to my parents, my wife Alexandra, my sister Anna, and my best friend Eldar Khalilov. Thank you all for your unconditional support, patience, and advice through all these years. Without you this thesis would not be possible.Finally, I would like to thank my friends from University of Waterloo who always supported me and made my studies much more joyful:

show abstract

Section: Subject Software Systemsmentioning

confidence: 99%

Transferring Pareto Frontiers across Heterogeneous Hardware Environments

Valov

Guo

Czarnecki

2020

Proceedings of the ACM/SPEC International Conference on Performance Engineering

View full text Add to dashboard Cite

show abstract

“…Learning from Synthetic Data is a promising direction in solving data scarcity issue, as the render engine could in theory produce unlimited number of synthetic data and their perfect annotations used for training. Many synthetic datasets have been released [56,43,14,32,5,8], for solving various pixel-level prediction tasks like semantic segmentation, optical flow, and the monocular depth prediction which is the focus of this paper. Note that a large body of work use the synthetic data to augment real-world datasets, which are already large in scale, to push the boundary of performance [33,8,50].…”

Section: Related Workmentioning

confidence: 99%

Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation

Zhao

Kong

Shin

et al. 2020

Preprint

View full text Add to dashboard Cite

Leveraging synthetically rendered data offers great potential to improve monocular depth estimation, but closing the synthetic-real domain gap is a non-trivial and important task. While much recent work has focused on unsupervised domain adaptation, we consider a more realistic scenario where a large amount of synthetic training data is supplemented by a small set of real images with groundtruth. In this setting we find that existing domain translation approaches are difficult to train and offer little advantage over simple baselines that use a mix of real and synthetic data. A key failure mode is that real-world images contain novel objects and clutter not present in synthetic training. This high-level domain shift isn't handled by existing image translation models.Based on these observations, we develop an attentional module that learns to identify and remove (hard) out-ofdomain regions in real images in order to improve depth prediction for a model trained primarily on synthetic data. We carry out extensive experiments to validate our attendremove-complete approach (ARC) and find that it significantly outperforms state-of-the-art domain adaptation methods for depth prediction. Visualizing the removed regions provides interpretable insights into the synthetic-real domain gap.

show abstract

“…In the next section, we show how an SLVS upload solution can make use of these properties to improve the QoE for all time-shifted viewers. Figure 1: SSIM as a function of the frame size for the Sintel trailer [10,26]. Each line corresponds to a single frame.…”

Section: Variability In the Upload Pathmentioning

confidence: 99%

“…The relationship between video bitrate and SSIM is typically observed to be a concave, non-decreasing function [39]. Figure 1 shows the relationship between the frame size and the SSIM for each frame of the Sintel [10,26] trailer. To keep the discussion in this section simple, we use a hypothetical model that reflects this relationship between the video bitrate (b) and the SSIM (Q).…”

Section: Inadequacy Of Existing Techniquesmentioning

confidence: 99%

Vantage

Ray

Kosaian

Rashmi

et al. 2019

Proceedings of the ACM Special Interest Group on Data Communication

View full text Add to dashboard Cite

Social live video streaming (SLVS) applications are becoming increasingly popular with the rise of platforms such as Facebook-Live, YouTube-Live, Twitch and Periscope. A key characteristic that differentiates this new class of applications from traditional live streaming is that these live streams are watched by viewers at different delays; while some viewers watch a live stream in real-time, others view the content in a time-shifted manner at different delays. In the presence of variability in the upload bandwidth, which is typical in mobile environments, existing solutions silo viewers into either receiving low latency video at a lower quality or a higher quality video with a significant delay penalty, without accounting for the presence of diverse time-shifted viewers. In this paper, we present Vantage, a live-streaming upload solution that improves the overall quality of experience for diverse time-shifted viewers by using selective quality-enhancing retransmissions in addition to real-time frames, optimizing the encoding schedules to balance the allocation of the available bandwidth between the two. Our evaluation using real-world mobile network traces shows that Vantage can provide high quality simultaneously for both low-latency and delayed viewing. For delayed viewing, Vantage achieves an average improvement of 19.9% over real-time optimized video streaming techniques across all the network traces and test videos, with observed gains of up to 42.9%. These benefits come at the cost of an average drop in realtime quality of 3.3%, with a maximum drop of 7.1%. This represents a significant performance improvement over current techniques used for SLVS applications, which primarily optimize the video upload for real-time viewing.

show abstract

Sintel

Cited by 5 publications

References 0 publications

Transferring Pareto Frontiers across Heterogeneous Hardware Environments

Transferring Pareto Frontiers across Heterogeneous Hardware Environments

Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation

Vantage

Contact Info

Product

Resources

About