Driven by advances in computer vision and the falling costs of camera hardware, organizations are deploying video cameras en masse for the spatial monitoring of their physical premises. Scaling video analytics to massive camera deployments, however, presents a new and mounting challenge, as compute cost grows proportionally to the number of camera feeds. This paper is driven by a simple question: can we scale video analytics in such a way that cost grows sublinearly, or even remains constant, as we deploy more cameras, while inference accuracy remains stable, or even improves. We believe the answer is yes. Our key observation is that video feeds from wide-area camera deployments demonstrate significant content correlations (e.g. to other geographically proximate feeds), both in space and over time. These spatio-temporal correlations can be harnessed to dramatically reduce the size of the inference search space, decreasing both workload and false positive rates in multi-camera video analytics. By discussing use-cases and technical challenges, we propose a roadmap for scaling video analytics to large camera networks, and outline a plan for its realization.
more »
« less
Video-zilla: An Indexing Layer for Large-Scale Video Analytics
Pervasive deployment of surveillance cameras today poses enormous scalability challenges to video analytics systems operating over many camera feeds. Currently, there are few indexing tools to organize video feeds beyond what is provided by a standard file system.
Recent video analytic systems implement application-specific frame profiling and sampling techniques to reduce the number of raw videos processed, leveraging frame-level redundancy or manually labeled spatial-temporal correlation between cameras. This paper presents Video-zilla, a standalone indexing layer between
video query systems and a video store to organize video data. We propose a video data unit abstraction, semantic video stream (SVS), based on a notion of distance between objects in the video. SVS implicitly captures scenes, which is missing from current video content characterization and a middle ground between individual
frames and an entire camera feed. We then build a hierarchical index that exposes the semantic similarity both within and across camera feeds, such that Video-zilla can quickly cluster video feeds based on their content semantics without manual labeling. We implement and evaluate Video-zilla in three use cases: object identification queries, clustering for training specialized DNNs, and archival services. In all three cases, Video-zilla reduces the time complexity
of inter-camera video analytics from linear with the number of cameras to sublinear, and reduces query resource usage by up to 14x compared to using frame-level or spatial-temporal similarity built into existing query systems.
more »
« less
- Award ID(s):
- 1815115
- NSF-PAR ID:
- 10393741
- Date Published:
- Journal Name:
- SIGMOD'22
- Page Range / eLocation ID:
- 1905 to 1919
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Cameras are deployed at scale with the purpose of searching and tracking objects of interest (e.g., a suspected person) through the camera network on live videos. Such cross-camera analytics is data and compute intensive, whose costs grow with the number of cameras and time. We present Spatula, a cost-efficient system that enables scaling cross-camera analytics on edge compute boxes to large camera networks by leveraging the spatial and temporal cross-camera correlations. While such correlations have been used in computer vision community, Spatula uses them to drastically reduce the communication and computation costs by pruning search space of a query identity (e.g., ignoring frames not correlated with the query identity’s current position). Spatula provides the first system substrate on which cross-camera analytics applications can be built to efficiently harness the cross-camera correlations that are abundant in large camera deployments. Spatula reduces compute load by $8.3\times$ on an 8-camera dataset, and by $23\times-86\times$ on two datasets with hundreds of cameras (simulated from real vehicle/pedestrian traces). We have also implemented Spatula on a testbed of 5 AWS DeepLens cameras.more » « less
-
Camera-based systems are increasingly used for collecting information on intersections and arterials. Unlike loop controllers that can generally be only used for detection and movement of vehicles, cameras can provide rich information about the traffic behavior. Vision-based frameworks for multiple-object detection, object tracking, and near-miss detection have been developed to derive this information. However, much of this work currently addresses processing videos offline. In this article, we propose an integrated two-stream convolutional networks architecture that performs real-time detection, tracking, and near-accident detection of road users in traffic video data. The two-stream model consists of a spatial stream network for object detection and a temporal stream network to leverage motion features for multiple-object tracking. We detect near-accidents by incorporating appearance features and motion features from these two networks. Further, we demonstrate that our approaches can be executed in real-time and at a frame rate that is higher than the video frame rate on a variety of videos collected from fisheye and overhead cameras.more » « less
-
It is a common practice to think of a video as a sequence of images (frames), and re-use deep neural network models that are trained only on images for similar analytics tasks on videos. In this paper, we show that this “leap of faith” that deep learning models that work well on images will also work well on videos is actually flawed.We show that even when a video camera is viewing a scene that is not changing in any humanperceptible way, and we control for external factors like video compression and environment (lighting), the accuracy of video analytics application fluctuates noticeably. These fluctuations occur because successive frames produced by the video camera may look similar visually, but are perceived quite differently by the video analytics applications.We observed that the root cause for these fluctuations is the dynamic camera parameter changes that a video camera automatically makes in order to capture and produce a visually pleasing video. The camera inadvertently acts as an “unintentional adversary” because these slight changes in the image pixel values in consecutive frames, as we show, have a noticeably adverse impact on the accuracy of insights from video analytics tasks that re-use image-trained deep learning models. To address this inadvertent adversarial effect from the camera, we explore the use of transfer learning techniques to improve learning in video analytics tasks through the transfer of knowledge from learning on image analytics tasks. Our experiments with a number of different cameras, and a variety of different video analytics tasks, show that the inadvertent adversarial effect from the camera can be noticeably offset by quickly re-training the deep learning models using transfer learning. In particular, we show that our newly trained Yolov5 model reduces fluctuation in object detection across frames, which leads to better tracking of objects (∼40% fewer mistakes in tracking). Our paper also provides new directions and techniques to mitigate the camera’s adversarial effect on deep learning models used for video analytics applications.more » « less
-
Online lecture videos are increasingly important e-learning materials for students. Automated content extraction from lecture videos facilitates information retrieval applications that improve access to the lecture material. A significant number of lecture videos include the speaker in the image. Speakers perform various semantically meaningful actions during the process of teaching. Among all the movements of the speaker, key actions such as writing or erasing potentially indicate important features directly related to the lecture content. In this paper, we present a methodology for lecture video content extraction using the speaker actions. Each lecture video is divided into small temporal units called action segments. Using a pose estimator, body and hands skeleton data are extracted and used to compute motion-based features describing each action segment. Then, the dominant speaker action of each of these segments is classified using Random forests and the motion-based features. With the temporal and spatial range of these actions, we implement an alternative way to draw key-frames of handwritten content from the video. In addition, for our fixed camera videos, we also use the skeleton data to compute a mask of the speaker writing locations for the subtraction of the background noise from the binarized key-frames. Our method has been tested on a publicly available lecture video dataset, and it shows reasonable recall and precision results, with a very good compression ratio which is better than previous methods based on content analysis.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3514221.3517840
