Recent advances in computer vision has led to a growth of interest in deploying visual analytics model on mobile devices. However, most mobile devices have limited computing power, which prohibits them from running large scale visual analytics neural networks. An emerging approach to solve this problem is to offload the computation of these neural networks to computing resources at an edge
server. Efficient computation offloading requires optimizing the trade-off between multiple objectives including compressed data rate, analytics performance, and computation
speed. In this work, we consider a “split computation” system to offload a part of the computation of the YOLO object detection model. We propose a learnable feature compression approach to compress the intermediate YOLO features with lightweight computation. We train the feature compression and decompression module together with the YOLO model to optimize the object detection accuracy
under a rate constraint. Compared to baseline methods that apply either standard image compression or learned image compression at the mobile and perform image de-compression and YOLO at the edge, the proposed system achieves higher detection accuracy at the low to medium rate range. Furthermore, the proposed system requires substantially lower computation time on the mobile device with CPU only.
more »
« less
Scalable Feature Compression for Edge-Assisted Object Detection Over Time-Varying Networks
Split-computing has recently emerged as a paradigm for offloading computation of visual analytics models from low-powered mobile devices to edge or cloud servers, by which the mobiles execute part of the model and compress and send the intermediate features, and the servers complete the remaining model computation. Prior feature compression approaches train different compression models and possibly visual analytics models to reach different target bit rates. We propose a scalable compression model that compresses the intermediate features of the YOLO object detection model into a layered bitstream, which can be easily adapted to meet the rate constraint of a dynamic network. Our approach achieves comparable rate-accuracy performance compared to prior non-scalable compression approaches over a large bitrate range.
more »
« less
- Award ID(s):
- 1952180
- PAR ID:
- 10466145
- Date Published:
- Journal Name:
- In MLSys Workshop on Resource-Constrained Learning in Wireless Networks
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Convolutional Neural Networks (CNN) have given rise to numerous visual analytics applications at the edge of the Internet. The image is typically captured by cameras and then live-streamed to edge servers for analytics due to the prohibitive cost of running CNN on computation-constrained end devices. A critical component to ensure low-latency and accurate visual analytics offloading over low bandwidth networks is image compression which minimizes the amount of visual data to offload and maximizes the decoding quality of salient pixels for analytics. Despite the wide adoption, JPEG standards and traditional image compression techniques do not address the accuracy of analytics tasks, leading to ineffective compression for visual analytics offloading. Although recent machine-centric image compression techniques leverage sophisticated neural network models or hardware architecture to support the accuracy-bandwidth trade-off, they introduce excessive latency in the visual analytics offloading pipeline. This paper presents CICO, a Context-aware Image Compression Optimization framework to achieve low-bandwidth and low-latency visual analytics offloading. CICO contextualizes image compression for offloading by employing easily-computable low-level image features to understand the importance of different image regions for a visual analytics task. Accordingly, CICO can optimize the trade-off between compression size and analytics accuracy. Extensive real-world experiments demonstrate that CICO reduces the bandwidth consumption of existing compression methods by up to 40% under comparable analytics accuracy. Regarding the low-latency support, CICO achieves up to a 2x speedup over state-of-the-art compression techniques.
-
Exploratory data analysis of high-dimensional datasets is a crucial task for which visual analytics can be especially useful. However, the ad hoc nature of exploratory analysis can also lead users to draw incorrect causal inferences. Previous studies have demonstrated this risk and shown that integrating counterfactual concepts within visual analytics systems can improve users’ understanding of visualized data. However, effectively leveraging counterfactual concepts can be challenging, with only bespoke implementations found in prior work. Moreover, it can require expertise in both counterfactual subset analysis and visualization to implement the functionalities practically. This paper aims to help address these challenges in two ways. First, we propose an operator-based conceptual model for the use of counterfactuals that is informed by prior work in visualization research. Second, we contribute the Co-op library, an open and extensible reference implementation of this model that can support the integration of counterfactual-based subset computation with visualization systems. To evaluate the effectiveness and generalizability of Co-op, the library was used to construct two different visual analytics systems each supporting a distinct user workflow. In addition, expert interviews were conducted with professional visual analytics researchers and engineers to gain more insights regarding how Co-op could be leveraged. Finally, informed in part by these evaluation results, we distil a set of key design implications for effectively leveraging counterfactuals in future visualization systems.more » « less
-
Many domains require analyst expertise to determine what patterns and data are interesting in a corpus. However, most analytics tools attempt to prequalify “interestingness” using algorithmic approaches to provide exploratory overviews. This overview-driven workflow precludes the use of qualitative analysis methodologies in large datasets. This paper discusses a preliminary visual analytics approach demonstrating how visual analytics tools can instead enable expert-driven qualitative analyses at scale by supporting computer-in-the-loop mixed-initiative approaches. We argue that visual analytics tools can support rich qualitative inference by using machine learning methods to continually model and refine what features correlate to an analyst’s on-going qualitative observations and by providing transparency into these features in order to aid analysts in navigating large corpora during qualitative analyses. We illustrate these ideas through an example from social media analysis and discuss open opportunities for designing visualizations that support qualitative inference through computer-in-the-loop approaches.more » « less
-
In Cloud 3D, such as Cloud Gaming and Cloud Virtual Reality (VR), image frames are rendered and compressed (encoded) in the cloud, and sent to the clients for users to view. For low latency and high image quality, fast, high compression rate, and high-quality image compression techniques are preferable. This paper explores computation time reduction techniques for learned image compression to make it more suitable for cloud 3D. More specifically, we employed slim (low-complexity) and application-specific AI models to reduce the computation time without degrading image quality. Our approach is based on two key insights: (1) as the frames generated by a 3D application are highly homogeneous, application-specific compression models can improve the rate-distortion performance over a general model; (2) many computer-generated frames from 3D applications are less complex than natural photos, which makes it feasible to reduce the model complexity to accelerate compression computation. We evaluated our models on six gaming image datasets. The results show that our approach has similar rate-distortion performance as a state-of-the-art learned image compression algorithm, while obtaining about 5x to 9x speedup and reducing the compression time to be less than 1 s (0.74s), bringing learned image compression closer to being viable for cloud 3D. Code is available at https://github.com/cloud-graphics-rendering/AppSpecificLIC.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
