1 code implementation • 12 May 2022 • S. H. Shabbeer Basha, Debapriya Tula, Sravan Kumar Vinakota, Shiv Ram Dubey
Later, in the second stage, the redundant filters are pruned from the fine-tuned CNN to decrease the network's complexity for the target task while preserving the performance.
1 code implementation • 26 Sep 2021 • Shiv Ram Dubey, S. H. Shabbeer Basha, Satish Kumar Singh, Bidyut Baran Chaudhuri
Overall, we observe very promising performance improvement of existing optimizers with the proposed AdaInject approach.
no code implementations • 30 Jan 2021 • S. H. Shabbeer Basha, Mohammad Farazuddin, Viswanath Pulabaigari, Shiv Ram Dubey, Snehasis Mukherjee
First, we train the model and select the filter pairs with redundant filters in each pair.
1 code implementation • 25 Apr 2020 • S. H. Shabbeer Basha, Sravan Kumar Vinakota, Viswanath Pulabaigari, Snehasis Mukherjee, Shiv Ram Dubey
The experimental results obtained in this study depict that tuning of the pre-trained CNN layers with the knowledge from the target dataset confesses better transfer learning ability.
no code implementations • 6 Feb 2020 • S. H. Shabbeer Basha, Viswanath Pulabaigari, Snehasis Mukherjee
Traditionally in deep learning based human activity recognition approaches, either a few random frames or every $k^{th}$ frame of the video is considered for training the 3D CNN, where $k$ is a small positive integer, like 4, 5, or 6.
no code implementations • 22 Jan 2020 • S. H. Shabbeer Basha, Sravan Kumar Vinakota, Shiv Ram Dubey, Viswanath Pulabaigari, Snehasis Mukherjee
Fine-tuning the newly learned (target-dependent) FC layers leads to state-of-the-art performance, according to the experiments carried out in this research.
1 code implementation • 21 Jan 2019 • S. H. Shabbeer Basha, Shiv Ram Dubey, Viswanath Pulabaigari, Snehasis Mukherjee
To automate the process of learning a CNN architecture, this paper attempts at finding the relationship between Fully Connected (FC) layers with some of the characteristics of the datasets.
1 code implementation • 30 Sep 2018 • S. H. Shabbeer Basha, Soumen Ghosh, Kancharagunta Kishan Babu, Shiv Ram Dubey, Viswanath Pulabaigari, Snehasis Mukherjee
The results of the proposed RCCNet model are compared with five state-of-the-art CNN models in terms of the accuracy, weighted average F1 score and training time.