FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3031774823> ?p ?o ?g. }

• W3031774823 abstract "When considering instances of distributed systems where visual sensors communicate with remote predictive models, data traffic is limited to the capacity of communication channels, and hardware limits the processing of collected data prior to transmission. We study novel methods of adapting visual inference to limitations on complexity and data availability at test time, wherever the aforementioned limitations exist. Our contributions detailed in this thesis consider both task-specific and task-generic approaches to reducing the data requirement for inference, and evaluate our proposed methods on a wide range of computer vision tasks. This thesis makes four distinct contributions: (i) We investigate multi-class action classification via two-stream convolutional neural networks that directly ingest information extracted from compressed video bitstreams. We show that selective access to macroblock motion vector information provides a good low-dimensional approximation of the underlying optical flow in visual sequences. (ii) We devise a bitstream cropping method by which AVC/H.264 and H.265 bitstreams are reduced to the minimum amount of necessary elements for optical flow extraction, while maintaining compliance with codec standards. We additionally study the effect of codec rate-quality control on the sparsity and noise incurred on optical flow derived from resulting bitstreams, and do so for multiple coding standards. (iii) We demonstrate degrees of variability in the amount of data required for action classification, and leverage this to reduce the dimensionality of input volumes by inferring the required temporal extent for accurate classification prior to processing via learnable machines. (iv) We extend the Mixtures-of-Experts (MoE) paradigm to adapt the data cost of inference for any set of constituent experts. We postulate that the minimum acceptable data cost of inference varies for different input space partitions, and consider mixtures where each expert is designed to meet a different set of constraints on input dimensionality. To take advantage of the flexibility of such mixtures in processing different input representations and modalities, we train biased gating functions such that experts requiring less information to make their inferences are favoured to others. We finally note that, our proposed data utility optimization solutions include a learnable component which considers specified priorities on the amount of information to be used prior to inference, and can be realized for any combination of tasks, modalities, and constraints on available data." @default.
• W3031774823 created "2020-06-05" @default.
• W3031774823 creator A5083380516 @default.
• W3031774823 date "2020-02-28" @default.
• W3031774823 modified "2023-09-26" @default.
• W3031774823 title "Adapting computer vision models to limitations on input dimensionality and model complexity" @default.
• W3031774823 cites W1005811612 @default.
• W3031774823 cites W1533861849 @default.
• W3031774823 cites W1549358575 @default.
• W3031774823 cites W1568437488 @default.
• W3031774823 cites W1663973292 @default.
• W3031774823 cites W1677182931 @default.
• W3031774823 cites W1686810756 @default.
• W3031774823 cites W1755205674 @default.
• W3031774823 cites W1849277567 @default.
• W3031774823 cites W1867429401 @default.
• W3031774823 cites W1947481528 @default.
• W3031774823 cites W1950136256 @default.
• W3031774823 cites W1956327946 @default.
• W3031774823 cites W1964168965 @default.
• W3031774823 cites W1969324951 @default.
• W3031774823 cites W1971735090 @default.
• W3031774823 cites W1985690171 @default.
• W3031774823 cites W2004915807 @default.
• W3031774823 cites W2008531776 @default.
• W3031774823 cites W2025653905 @default.
• W3031774823 cites W2027902935 @default.
• W3031774823 cites W2028166238 @default.
• W3031774823 cites W2032756980 @default.
• W3031774823 cites W2042970394 @default.
• W3031774823 cites W2044535169 @default.
• W3031774823 cites W2046811824 @default.
• W3031774823 cites W2050576295 @default.
• W3031774823 cites W2053816989 @default.
• W3031774823 cites W2056100347 @default.
• W3031774823 cites W2062227835 @default.
• W3031774823 cites W2074130315 @default.
• W3031774823 cites W2080531309 @default.
• W3031774823 cites W2081614366 @default.
• W3031774823 cites W2084910356 @default.
• W3031774823 cites W2093383096 @default.
• W3031774823 cites W2094498958 @default.
• W3031774823 cites W2100495367 @default.
• W3031774823 cites W2101095383 @default.
• W3031774823 cites W2105658140 @default.
• W3031774823 cites W2108598243 @default.
• W3031774823 cites W2109703216 @default.
• W3031774823 cites W2111719156 @default.
• W3031774823 cites W2115122836 @default.
• W3031774823 cites W2117496083 @default.
• W3031774823 cites W2117539524 @default.
• W3031774823 cites W2118877769 @default.
• W3031774823 cites W2119044429 @default.
• W3031774823 cites W2122825543 @default.
• W3031774823 cites W2126579184 @default.
• W3031774823 cites W2131111770 @default.
• W3031774823 cites W2133081131 @default.
• W3031774823 cites W2133665775 @default.
• W3031774823 cites W2133728753 @default.
• W3031774823 cites W2134053106 @default.
• W3031774823 cites W2136836265 @default.
• W3031774823 cites W2137983211 @default.
• W3031774823 cites W2139920000 @default.
• W3031774823 cites W2146395539 @default.
• W3031774823 cites W2149684865 @default.
• W3031774823 cites W2150884987 @default.
• W3031774823 cites W2151162785 @default.
• W3031774823 cites W2152435897 @default.
• W3031774823 cites W2152839228 @default.
• W3031774823 cites W2156303437 @default.
• W3031774823 cites W2158401301 @default.
• W3031774823 cites W2159505618 @default.
• W3031774823 cites W2163605009 @default.
• W3031774823 cites W2167050501 @default.
• W3031774823 cites W2172654076 @default.
• W3031774823 cites W2173520492 @default.
• W3031774823 cites W2187089797 @default.
• W3031774823 cites W2187230075 @default.
• W3031774823 cites W2188183693 @default.
• W3031774823 cites W2194775991 @default.
• W3031774823 cites W2209882149 @default.
• W3031774823 cites W2272300165 @default.
• W3031774823 cites W2342662179 @default.
• W3031774823 cites W2401760721 @default.
• W3031774823 cites W24089286 @default.
• W3031774823 cites W2476548250 @default.
• W3031774823 cites W2479750863 @default.
• W3031774823 cites W2503339013 @default.
• W3031774823 cites W2507536322 @default.
• W3031774823 cites W2530345236 @default.
• W3031774823 cites W2534320940 @default.
• W3031774823 cites W2550553598 @default.
• W3031774823 cites W2553303224 @default.
• W3031774823 cites W2557728737 @default.
• W3031774823 cites W2560474170 @default.
• W3031774823 cites W2586564970 @default.
• W3031774823 cites W2593768305 @default.
• W3031774823 cites W2604392022 @default.
• W3031774823 cites W2605488490 @default.
• W3031774823 cites W2608988379 @default.

• next