9 open source projects by val-iisc

This project is an implementation of the crowd counting model proposed in our CVPR 2017 paper - Switching Convolutional Neural Network(SCNN) for Crowd Counting. SCNN is an adaptation of the fully-convolutional neural network and uses an expert CNN that chooses the best crowd density CNN regressor for parts of the scene from a bag of regressors. This helps it tackle intra-scene crowd density variation and obtain SOTA results

This project is an implementation of the Generative Adversarial Network proposed in our CVPR 2017 paper - DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data. DeLiGAN is a simple but effective modification of the GAN framework and aims to improve performance on datasets which are diverse yet small in size.

Visualising predictions of deep neural networks

Generalized Data-free Universal Adversarial Perturbations

[CVPR 2018] Tensorflow implementation of NAG : Network for Adversary Generation

Code release for "From Image Collections to Point Clouds with Self-supervised Shape and Pose Networks" (CVPR 2020)

Repository for 'Dense 3D Point Cloud Reconstruction Using a Deep Pyramid Network' [WACV 2019]

Data independent universal adversarial perturbations