Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Deep neural networks for video classification in ecology
Analyzing large volumes of video data is a challenging and time-consuming task. Automating this process would very valuable, especially in ecological research where massive amounts of video can be used to unlock new avenues of ecological research into the behaviour of animals in their environments....
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
University of Cape Town
2021
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613813517320192 |
|---|---|
| access_status_str | Open Access |
| author | Conway, Alexander |
| author2 | Durbach, Ian |
| author_browse | Conway, Alexander Durbach, Ian |
| author_facet | Durbach, Ian Conway, Alexander |
| author_sort | Conway, Alexander |
| collection | Thesis |
| description | Analyzing large volumes of video data is a challenging and time-consuming task. Automating this process would very valuable, especially in ecological research where massive amounts of video can be used to unlock new avenues of ecological research into the behaviour of animals in their environments. Deep Neural Networks, particularly Deep Convolutional Neural Networks, are a powerful class of models for computer vision. When combined with Recurrent Neural Networks, Deep Convolutional models can be applied to video for frame level video classification. This research studies two datasets: penguins and seals. The purpose of the research is to compare the performance of image-only CNNs, which treat each frame of a video independently, against a combined CNN-RNN approach; and to assess whether incorporating the motion information in the temporal aspect of video improves the accuracy of classifications in these two datasets. Video and image-only models offer similar out-of-sample performance on the simpler seals dataset but the video model led to moderate performance improvements on the more complex penguin action recognition dataset. |
| format | Thesis |
| id | oai:open.uct.ac.za:11427/32520 |
| institution | University of Cape Town (South Africa) |
| language | eng |
| last_indexed | 2026-06-10T12:42:06.917Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UCTD — University of Cape Town Open Access Repository |
| publishDate | 2021 |
| publishDateRange | 2021 |
| publishDateSort | 2021 |
| publisher | University of Cape Town |
| publisherStr | University of Cape Town |
| record_format | dspace |
| source_str | UCTD — University of Cape Town Open Access Repository |
| spelling | oai:open.uct.ac.za:11427/32520 Deep neural networks for video classification in ecology Conway, Alexander Durbach, Ian Deep Neural Networks Computer Vision Analyzing large volumes of video data is a challenging and time-consuming task. Automating this process would very valuable, especially in ecological research where massive amounts of video can be used to unlock new avenues of ecological research into the behaviour of animals in their environments. Deep Neural Networks, particularly Deep Convolutional Neural Networks, are a powerful class of models for computer vision. When combined with Recurrent Neural Networks, Deep Convolutional models can be applied to video for frame level video classification. This research studies two datasets: penguins and seals. The purpose of the research is to compare the performance of image-only CNNs, which treat each frame of a video independently, against a combined CNN-RNN approach; and to assess whether incorporating the motion information in the temporal aspect of video improves the accuracy of classifications in these two datasets. Video and image-only models offer similar out-of-sample performance on the simpler seals dataset but the video model led to moderate performance improvements on the more complex penguin action recognition dataset. 2021-01-15T09:53:08Z 2021-01-15T09:53:08Z 2020 Master Thesis Masters MSc http://hdl.handle.net/11427/32520 eng application/pdf University of Cape Town Department of Statistical Sciences Faculty of Science |
| spellingShingle | Deep Neural Networks Computer Vision Conway, Alexander Deep neural networks for video classification in ecology |
| thesis_degree_str | Master's |
| title | Deep neural networks for video classification in ecology |
| title_full | Deep neural networks for video classification in ecology |
| title_fullStr | Deep neural networks for video classification in ecology |
| title_full_unstemmed | Deep neural networks for video classification in ecology |
| title_short | Deep neural networks for video classification in ecology |
| title_sort | deep neural networks for video classification in ecology |
| topic | Deep Neural Networks Computer Vision |
| url | http://hdl.handle.net/11427/32520 |
| work_keys_str_mv | AT conwayalexander deepneuralnetworksforvideoclassificationinecology |