Pinned Loading

1. Constructing-Per-Shot-Bitrate-Ladders-using-Visual-Information-Fidelity Constructing-Per-Shot-Bitrate-Ladders-using-Visual-Information-Fidelity Public

   We develop a perceptually optimized method of constructing optimal per-shot bitrate and quality ladders, using an ensemble of low-level features and Visual Information Fidelity (VIF) features extra…

   Python

2. AutoML-Models-for-Wireless-Signals-Classification-and-effectiveness-against-Adversarial-Attacks AutoML-Models-for-Wireless-Signals-Classification-and-effectiveness-against-Adversarial-Attacks Public

   Comparing and understanding the performance of AutoML models with state-of-the-art models on wireless signal classification and their vulnerability towards transfer-based Projected Gradient Descent…

   Jupyter Notebook 5 2

3. SR-DDPM SR-DDPM Public

   Forked from shreshthsaini/SR-DDPM

   Denoising diffusion probabilistic model for low level vision task. Developing a novel DM for super resolution task; later to be extended for general vision tasks such as deblurring, dehazing, rain …

   Jupyter Notebook

4. Optical-Flow-Less-Video-Frame-Interpolation Optical-Flow-Less-Video-Frame-Interpolation Public

   A modified light weight VRT is used to predict intermediate frames by looking only the previous frames or following causality without any use optical flow estimation techniques.

   Python

5. White-Box-Cartoonization White-Box-Cartoonization Public

   Experiments with two different models mainly VGG19 and VIT-16 which are used for calculating Structure and Content Losses in White Box Cartoonization and understanding the performance of the cartoo…

   Python 1

6. Effects-of-reduced-frame-corruptions-on-video-classification Effects-of-reduced-frame-corruptions-on-video-classification Public

   Exploring the vulnerability of CNN-RNN based video classification model towards reduced frames natural and adversarial perturbations.

   Jupyter Notebook 1