Visual Sentiment Analysis

ADE20K Dataset

Contrastive Language-Image Pretraining

Real-time face detection and emotion/gender classification using fer2013/imdb datasets with a keras CNN model and openCV.

State-of-the-art deep learning model for analyzing sentiment, emotion, sarcasm etc.

DeViSE model (zero-shot learning) trained on ImageNet and deployed on AWS using Docker

One-click image sorting/labelling script

A CV toolkit for my papers.

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

Support, annotation, evaluation, and baseline models for the imSitu dataset.

Code from the paper "Step-by-Step: Separating Planning from Realization in Neural Data-to-Text Generation - NAACL-2019.

Code for the paper "Language Models are Unsupervised Multitask Learners"

The Places365-CNNs for Scene Classification

Incubator for useful bioinformatics code, primarily in Python and R

Tensorflow code for CVPR 2017 paper: Learning a Deep Embedding Model for Zero-Shot Learning

ImageNet image downloader

Development kit for MIT Scene Parsing Benchmark

EmotionNet2 convolutional neural network for expression recognition