🔥 CNN for Watermark Removal using Deep Image Prior with Pytorch 🔥.

Segmentation pipeline that uses a U-Net backbone to perform segmentation on the Cityscapes dataset. Conducted experiments to analyse the impact of the skip connections of the U-Net on the quality of the segmentation masks. These masks are also qualitatively analysed using the Intersection-over-Union (IoU) metric

Nested U-Net with two-level skip connections for speech enhancement

This project demonstrates the implementation of ResNet50 from scratch and its application for chest cancer classification using the Chest CT-scan Images dataset.

ResNet-34 Model trained from scratch to classify 450 different species of birds with 98.6% accuracy.

The main goal of this project is to come up with an architecture having the highest test accuracy on the CIFAR-10 image classification dataset, under the constraint that model has no more than 5 million parameters.

Implementation of Yolo v3 object detection fully convolutional neural network model in Tensorflow & Keras

Code accompanying the paper: "Hybrid Skip: A Biologically Inspired Skip Connection for the UNet Architecture"

PyTorch Implementation of Hybrid Skip Connection for UNet

road and traffic segmentation with IoU metric and DICE coffecient

This repository contains the implementation of famous ResNet50 which is state of art technique for image classification

IEEE paper implementation of Single-View 2D-3D Reconstruction.

[TPAMI 2022] "Bag of Tricks for Training Deeper Graph Neural Networks A Comprehensive Benchmark Study" by Tianlong Chen*, Kaixiong Zhou*, Keyu Duan, Wenqing Zheng, Peihao Wang, Xia Hu, Zhangyang Wang

Model Pipelines for GNNs, VAEs, Neural Style Transfer, and other kinds of models!

The project presents a comparative study of Brain Tumor Segmentation using 3 approaches - 1) Sobel Operator and U-Net, 2) V-Net, 3) W-Net

Source code (train/test) accompanying the paper entitled "Veritatem Dies Aperit - Temporally Consistent Depth Prediction Enabled by a Multi-Task Geometric and Semantic Scene Understanding Approach" in CVPR 2019 (https://arxiv.org/abs/1903.10764).

This is the official implementation of "Novel View Synthesis with Skip Connections" (ICIP 2020)

Codes for ICLR 2020 paper "Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNets"

Reconstructing Medical Images using Generative model.

Replication of Jasper speech-to-text network using Intel optimized TensorFlow.