Content for workshops on image dataset curation @ HPI's AI Service Center

iSeeBetter: Spatio-Temporal Video Super Resolution using Recurrent-Generative Back-Projection Networks | Python3 | PyTorch | GANs | CNNs | ResNets | RNNs | Published in Springer Journal of Computational Visual Media, September 2020, Tsinghua University Press

Journey to Learn Deep Learning with Pytorch from scratch i.e, from Tensor & Gradients to Advance topic like Generative Adversarial Networks

This Project uses Convolutional Neural Networks (CNN) for the classification and prediction of handwritten Devanagari script. Leveraging transfer learning techniques, it adapts pre-trained models to recognize and forecast characters in Devanagari, enhancing accuracy and efficiency.

Implementation of "Weight Averaging Improves Knowledge Distillation under Domain Shift" (ICCV 2023 OOD-CV Workshop)

Code for paper "EdgeKE: An On-Demand Deep Learning IoT System for Cognitive Big Data on Industrial Edge Devices"

A set of experiments inspired by the paper "Training BatchNorm and Only BatchNorm: On the Expressive Power of Random Features in CNNs" by Jonathan Frankle, David J. Schwab, Ari S. Morcos

Training using an alternative approach: forward-thinking

Robustness of Deep Neural Networks using Trainable Activation Functions

Yolov3 with multi backbone

Designed a smaller architecture implemented from the paper Deep Residual Learning for Image Recognition and achieved 93.65% accuracy.

Implemented Deep Residual Learning for Image Recognition Paper and achieved lower error rate by customizing different parts of the architecture.

Implemented the Deep Residual Learning for Image Recognition Paper and achieved better accuracy by customizing different parts of the architecture.

Optimize ResNet Learning Process

Pytorch Implementation and Performance Analysis of the Popular Vision Architectures from Scratch.

Single-sequence and Profile-based Prediction of RNA Solvent Accessibility Using Dilated Convolution Neural Network

High Accuracy ResNet Model under 5 Million parameters.

Code for DCASE 2020 task 1a and task 1b.

A collection of small-scale projects that helped me learn the basics of the PyTorch framework