CoChat

NoScope: optimizing neural network queries over video at scale

Open access
Kang, Daniel, Emmons, John, Abuzaid, Firas, Bailis, Peter, Zaharia, Matei · 2017 · Proceedings of the VLDB Endowment

The DOI matches its Crossref record; 102 of the 102 references it cites were found in a registry or an index.

Abstract

Recent advances in computer vision---in the form of deep neural networks---have made it possible to query increasing volumes of video data with high accuracy. However, neural network inference is computationally expensive at scale: applying a state-of-the-art object detector in real time (i.e., 30+ frames per second) to a single video requires a $4000 GPU. In response, we present N o S cope , a system for querying videos that can reduce the cost of neural network video analysis by up to three orders of magnitude via inference-optimized model search. Given a target video, object to detect, and reference neural network, N o S cope automatically searches for and trains a sequence, or cascade, of models that preserves the accuracy of the reference network but is specialized to the target video and are therefore far less computationally expensive. N o S cope cascades two types of models: specialized models that forego the full generality of the reference model but faithfully mimic its behavior for the target video and object; and difference detectors that highlight temporal differences across frames. We show that the optimal cascade architecture differs across videos and objects, so N o S cope uses an efficient cost-based optimizer to search across models and cascades. With this approach, N o S cope achieves two to three order of magnitude speed-ups (265-15,500x real-time) on binary classification tasks over fixed-angle webcam and surveillance video while maintaining accuracy within 1--5% of state-of-the-art neural networks.

Overview

Integrity signals

Registry

Crossref · 10.14778/3137628.3137664

Cites

102 of 102 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Citations

Cites

Research works printed but not linked: 24

  1. Abadi D. J., 2005, CIDR
  2. Ba J., 2014, NIPS
  3. Bailis P., 2017, CIDR
  4. Benenson R., 2014, ECCV
  5. Chandrasekaran S., 2003, CIDR
  6. Chaudhuri S., 2007, VLDB
  7. Chen W., 2015, ICML
  8. Feichtenhofer C., 2016, NIPS
  9. 10.5555/2354409.2354978 · doi:10.5555/2354409.2354978
  10. Gibbs S., 1993, ICDE
  11. Goodfellow I., 2016, MIT Press
  12. Han S., 2016, ICLR
  13. Hinton G., 2014, NIPS
  14. Idreos S., 2007, CIDR
  15. Kraska T., 2013, CIDR
  16. Krizhevsky A., 2012, NIPS
  17. 10.5555/850924.851523 · doi:10.5555/850924.851523
  18. Metz C., 2017, Wired
  19. Patait A., 2016, 2016 GPU Technology Conference (https://goo.gl/Bdjdgm)
  20. Ren S., 2015, NIPS
  21. Yue-Hei Ng J., 2015, CVPR, 53
  22. Zhang H., 2017, NSDI
  23. Zhanget S., 2016, CVPR
  24. Zoph B., 2017, ICLR

Cited by

See all 469

Metadata

Authors
Kang, D., Emmons, J., Abuzaid, F., Bailis, P., & Zaharia, M.
Affiliations
Stanford University (US)
Venue
Proceedings of the VLDB Endowment
Published
2017-08
Volume
10
Issue
11
Pages
1586-1597
ISSN
2150-8097 (ISSN-L)
OpenAlex
W2752236330
Cited by
367
References
97
Citations by year
2026: 23 · 2025: 33 · 2024: 35 · 2023: 40 · 2022: 60 · 2021: 49 · 2020: 45 · 2019: 53 · 2018: 28 · 2017: 1
FWCI
14.42
Type
journal article
Language
en
Open access
Licence
cc-by-sa
Locations
Proceedings of the VLDB Endowment (publishedVersion, cc-by-sa)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Surveillance and Tracking Methods
  • Physical Sciences › Computer Science › Information Systems › Advanced Database Systems and Queries
Keywords
object detection, inference optimization, model search, cascade model, cost-based optimization, real-time video analysis, surveillance video, webcam video, binary classification

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.