Dr Bappaditya Mandal's Outputs (5)

DeepPCA Based Objective Function for Melanoma Detection (2018)
Presentation / Conference Contribution
Sultana, N. N., Puhan, N. B., & Mandal, B. (2018, December). DeepPCA Based Objective Function for Melanoma Detection. Presented at 2018 International Conference on Information Technology (ICIT), Bhubaneswar, India

In this paper, we propose an objective function for the convolutional neural network to acquire the variation separability as opposed to the categorical cross entropy which maximizes according to the target labels. This approach is an unsupervised le... Read More about DeepPCA Based Objective Function for Melanoma Detection.

Deep Residual Network with Regularized Fisher Framework for Detection of Melanoma (2018)
Journal Article
Mandal, B. (2018). Deep Residual Network with Regularized Fisher Framework for Detection of Melanoma. IET Computer Vision, 1096-1104. https://doi.org/10.1049/iet-cvi.2018.5238

Of all the skin cancer that is prevalent, melanoma has the highest mortality rates. Melanoma becomes life threatening when it penetrates deep into the dermis layer unless detected at an early stage, it becomes fatal since it has a tendency to migrate... Read More about Deep Residual Network with Regularized Fisher Framework for Detection of Melanoma.

Deep convolutional neural network-based patch classification for retinal nerve fiber layer defect detection in early glaucoma (2018)
Journal Article
Panda, R., Puhan, N. B., Rao, A., Mandal, B., Padhy, D., & Panda, G. (2018). Deep convolutional neural network-based patch classification for retinal nerve fiber layer defect detection in early glaucoma. Journal of Medical Imaging, 5(04), 1. https://doi.org/10.1117/1.jmi.5.4.044003

Deep Adaptive Temporal Pooling for Activity Recognition (2018)
Presentation / Conference Contribution
Song, S., Cheung, N.-M., Chandrasekhar, V., & Mandal, B. (2018, October). Deep Adaptive Temporal Pooling for Activity Recognition. Presented at MM '18: ACM Multimedia Conference, Seoul Republic of Korea

Deep neural networks have recently achieved competitive accuracy for human activity recognition. However, there is room for improvement, especially in modeling of long-term temporal importance and determining the activity relevance of different tempo... Read More about Deep Adaptive Temporal Pooling for Activity Recognition.

Deep Residual Network With Subclass Discriminant Analysis For Crowd Behavior Recognition (2018)
Presentation / Conference Contribution
Mandal, B., Fajtl, J., Argyriou, V., Monekosso, D., & Remagnino, P. (2018, October). Deep Residual Network With Subclass Discriminant Analysis For Crowd Behavior Recognition. Presented at 2018 IEEE International Conference on Image Processing, Athens

In this work, we extract rich representations of crowd behavior from video using a fine-tuned deep convolutional neural residual network. Using spatial partitioning trees we create subclasses within the feature maps from each of the crowd behavior a... Read More about Deep Residual Network With Subclass Discriminant Analysis For Crowd Behavior Recognition.