no code implementations • 7 Nov 2023 • Bipul Neupane, Jagannath Aryal, Abbas Rajabifard
The workflow first identifies the best (with the highest evaluation scores) hyperparameters, lightweight CNNs for the Student (among 43 CNNs from Computer Vision), and encoder-decoder networks (EDNs) for both Teachers and Students.
no code implementations • 27 Aug 2023 • Chiranjibi Sitaula, Jagannath Aryal, Avik Bhattacharya
Classification of very high-resolution (VHR) aerial remote sensing (RS) images is a well-established research area in the remote sensing community as it provides valuable spatial information for decision-making.
no code implementations • 17 Jun 2023 • Aditya Aditya, Bharat Lohani, Jagannath Aryal, Stephan Winter
PointMetaBase and KPConv (omni-supervised) achieve the highest mIoU on the Chandigarh (95. 24%) and Toronto3D datasets (91. 26%), respectively while PointCNN provides the highest mIoU on the Kerala dataset (85. 68%).
no code implementations • 8 May 2023 • Yuanzhi Cai, Jagannath Aryal, Yuan Fang, Hong Huang, Lei Fan
In this study, the concept of pruning from a supernet is used for the first time to integrate the selection of channel combination and the training of a semantic segmentation network.
no code implementations • 1 May 2023 • Chiranjibi Sitaula, Sumesh KC, Jagannath Aryal
Very high-resolution (VHR) remote sensing (RS) scene classification is a challenging task due to the higher inter-class similarity and intra-class variability problems.
1 code implementation • 16 Mar 2023 • Bipul Neupane, Jagannath Aryal, Abbas Rajabifard
The results conclude that selectively densifying feature maps and skip connections enhances network performance without a substantial increase in parameters.
no code implementations • 15 Jun 2022 • Chiranjibi Sitaula, Tej Bahadur Shahi, Faezeh Marzbanrad, Jagannath Aryal
For this, we, first, devise the taxonomy using the seminal existing methods proposed in the literature to this date {using deep learning (DL)-based, computer vision (CV)-based, and search engine (SE)-based methods}.