A Technical Talk on “Deep Learning for Image Recognition”

The department of Computer Science and Engineering in coordination with the Research review cell organized a technical talk on “Deep Learning for Image Recognition” through virtual mode for the faculty and student community of the SMVITM, Bantakal on 9^th June 2021 at 4:00 PM to 5:45 PM. The resource person of the talk is Dr. Sachin Bhat, Associate Professor, Dept. of ECE, SMVITM.

Dr. Sachin began the presentation by providing a brief history of AI, machine learning, and deep learning, before highlighting the deep learning taxonomy. Dr. Bhat stressed the importance of neural networks. A neural network is a network or circuit of neurons, or, in a modern sense, an artificial neural network made of artificial neurons or nodes.

Dr.  Bhat also highlighted some issues while taking the input as an image and the use of Convolutional neural networks, along with DCNN and its categories. During his talk, he also demonstrated the usage of Google Colab to build a simple CNN for Devanagari text categorization, with 46 classes consisting of the fundamental alphabets and numbers and discussed the pseudocode for training the CNN model. Feature maps, activation functions, optimization, and hyperparameters were all discussed by Dr. Sachin. Finally, Dr. Bhat presented the findings of the ILSVRC investigation comparing deep learning to human levels of accuracy. The session ended with an enthusiastic question-answer round from the audience and the orator. Ms. Rukmini Bhat B, Assistant Professor, gave a warm welcome to the audience at the beginning of the event. She has also introduced the resource person. She has concluded the webinar by thanking all attendees and expressed her gratitude at the end of the program.