Search

Search Funnelback University

Search powered by Funnelback
1 - 6 of 6 search results for KaKaoTalk:PC53 24 / |u:mobile-systems.cl.cam.ac.uk where 0 match all words and 6 match some words.
  1. Results that match 1 of 2 words

  2. Accelerating Mobile Audio Sensing Algorithmsthrough On-Chip GPU…

    https://mobile-systems.cl.cam.ac.uk/papers/mobisys17.pdf
    10 May 2017: As aresult, computational offloading to cloud [24] or low-power co-processors [27, 45] has often been the solution applied to keep theseapps functional on the mobile device, but no study
  3. A Study of Bluetooth Low Energy Performance forHuman Proximity ...

    https://mobile-systems.cl.cam.ac.uk/papers/percom17.pdf
    11 Jan 2017: A. BLE Modes of Operation. BLE provides two modes of communication: connectionbased and broadcast based [24].
  4. Understanding the Role of Places and Activities on Mobile Phone…

    https://mobile-systems.cl.cam.ac.uk/papers/Ubicomp17-Abhinav.pdf
    1 Aug 2017: 24, 30] and how these are inuenced by context [28], content [23], and the complexity of an ongoingtask [24]. ... Previous studies have exploredvarious aspects of users’ interaction with mobile notications [4, 6, 7, 12, 24, 29, 35].
  5. UbiComp17_camera

    https://mobile-systems.cl.cam.ac.uk/papers/Ubicomp17-Georgiev.pdf
    1 Aug 2017: Georgiev et al. 1 INTRODUCTIONFor a wide range of sensory perception tasks, current state-of-the-art techniques rely on various forms of deeplearning; typical examples include: recognizing an object [24] ... We use the LITIS Rouen Audio Scene dataset
  6. Multimodal Deep Learning for Activity and Context Recognition

    https://mobile-systems.cl.cam.ac.uk/papers/ubicomp2018-radu.pdf
    21 Dec 2017: Feature Concatenation: Deep vs. Shallow. Considering both shallow and deep classifiers adopting FC, FC-deepclassifiers on average outperform FC-shallow classifiers, by 24% in many cases.
  7. SIGCHI Conference Proceedings Format

    https://mobile-systems.cl.cam.ac.uk/papers/www17.pdf
    17 Feb 2017: at university and on full-time/self em-ployment are usually the most active (0.25, 0.24 and 0.23). ... MoodScope [34]takes an orthogonal approach and uses smartphone usage patterns,for example, browsing history, phone calls, to infer the user’s

Refine your results