Junwei Liang

We propose an end-to-end, multi-task learning system utilizing rich visual features about human behavioral information and interaction with their surroundings. Expand

We propose a novel neural network called Focal Visual-Text Attention network (FVTA) for collective reasoning in visual question answering, where both visual and text sequence information such as images and text metadata are presented. Expand

This paper proposes a new task, MemexQA: given a collection of photos or videos from a user, the goal is to automatically answer questions that help users recover their memory about events captured in the collection. Expand

This paper studies the problem of predicting the distribution over multiple possible future paths of people as they move through various visual scenes. Expand

We present an event detection system, which shares many similarities with standard object detection pipelines. Expand

We propose an end-to-end, multi-task learning system utilizing rich visual features about human behavioral information and interaction with their surroundings to predict a pedestrian's future path jointly with future activities. Expand

This paper proposes a new multimodal MemexQA task: given a sequence of photos from a user, the goal is to automatically answer questions that help users recover their memory about an event captured in these photos. Expand

In this paper, we study the problem of efficiently assessing building damage after natural disasters like hurricanes, floods or fires, through aerial video analysis. Expand

The recent advances in image captioning stimulate the research in generating natural language description for visual content, which can be widely applied in many applications. Expand