In object detection, an intersection over union (IoU) threshold is required to define positives
and negatives. An object detector, trained with low IoU threshold, eg 0.5, usually produces …

Top-down visual attention mechanisms have been used extensively in image captioning
and visual question answering (VQA) to enable deeper image understanding through fine …

Problems at the intersection of vision and language are of significant importance both as
challenging research questions and for the rich set of applications they enable. However …

We propose a technique for producing'visual explanations' for decisions from a large class
of Convolutional Neural Network (CNN)-based models, making them more transparent. Our …

We propose a technique for producing 'visual explanations' for decisions from a large class
of Convolutional Neural Network (CNN)-based models, making them more transparent and …

The multimodal task of Visual Question Answering (VQA) encompassing elements of
Computer Vision (CV) and Natural Language Processing (NLP), aims to generate answers …

We introduce the task of Visual Dialog, which requires an AI agent to hold a meaningful
dialog with humans in natural, conversational language about visual content. Specifically …

A number of studies have found that today's Visual Question Answering (VQA) models are
heavily driven by superficial correlations in the training data and lack sufficient image …

We propose a technique for making Convolutional Neural Network (CNN)-based models
more transparent by visualizing input regions that are'important'for predictions--or visual …

This paper presents a state-of-the-art model for visual question answering (VQA), which won
the first place in the 2017 VQA Challenge. VQA is a task of significant importance for …