saumya-jetley

causality

Heart Attack: Causality Challenge

cd_Adversaries_in_attention

cd_Flag_recognition

Recognition of country flags as found in real-world images

cd_ICLR18_LearnToPayAttention

Active attention in classification networks that is optimised at the time of model training.

cd_mean_shift

mean-shift code

cd_PoseAnalysis_PASCAL3D-STS

Nearest-neighbour pose analysis in the downsamples binary, radial and learned embedding spaces established for the shape silhouettes of the dataset.

cd_Prototypical_Priors_BMVC15

The code for BMVC15 submission - leveraging prototypical priors for classification to zero shot learning

cd_Saliency_PDP

Saliency estimation as probability distribution prediction - Implementation of the CVPR'16 spotlight paper by the same name.

straighttoshapes

Straight To Shapes: Real-Time Detection of Encoded Shapes

whoneedsadversaries

Code implementing the experiments described in the NeurIPS 2018 paper "With Friends Like These, Who Needs Adversaries?".

saumya-jetley/cd_ICLR18_LearnToPayAttention

Active attention in classification networks that is optimised at the time of model training.

saumya-jetley/cd_Saliency_PDP

Saliency estimation as probability distribution prediction - Implementation of the CVPR'16 spotlight paper by the same name.

saumya-jetley/causality

Heart Attack: Causality Challenge

saumya-jetley/cd_Adversaries_in_attention

saumya-jetley/cd_Flag_recognition

Recognition of country flags as found in real-world images

saumya-jetley/cd_mean_shift

mean-shift code

saumya-jetley/cd_PoseAnalysis_PASCAL3D-STS

Nearest-neighbour pose analysis in the downsamples binary, radial and learned embedding spaces established for the shape silhouettes of the dataset.

saumya-jetley/cd_Prototypical_Priors_BMVC15

The code for BMVC15 submission - leveraging prototypical priors for classification to zero shot learning

saumya-jetley/cd_SketchIt

Auto sketch artist: Sketches an edge map of the input image interactively

saumya-jetley/cd_ZeroShotSeg_cocomAPI

COCO Matlab API repurposed to evaluate zero-shot segmentation performance on unseen categories i.e. the 60 COCO categories that are not in common with the 20 categories of the PASCAL dataset.

saumya-jetley/cifar.torch

92.45% on CIFAR-10 in Torch

saumya-jetley/dpnn

deep extensions to nn

saumya-jetley/kmeans-learning-torch

K-means feature learning on CIFAR-10 translated to Torch

saumya-jetley/nn

saumya-jetley/nngraph

Graph Computation for nn

saumya-jetley/openslide-matlab

A MATLAB library providing bindings to the C library OpenSlide (https://github.com/openslide/openslide) used for reading whole-slide images.

saumya-jetley/phd-thesis-template

PhD dissertation template in LaTeX

saumya-jetley/pp_ICVSS15_TacklingBkgndDifferently

Classifying background using a learned threshold. This affords smooth incorporation of attribute-based embedding space over the non-background category labels in classification models.

saumya-jetley/saumya-jetley.github.io

Personal website (template rights belong to blackrock digital)

saumya-jetley/torch-toolbox-ADV

A collection of snippets and libraries for Torch from e-Lab

saumya-jetley/torch7

http://torch.ch