Scene Graph Prediction with Limited Labels

Authors Vincent Chen, Paroma Varma, Ranjay Krishna, Michael Bernstein, Christopher Re, Li Fei-Fei Our semi-supervised method automatically generates probabilistic relationship labels to train any scene graph model. Abstract Visual knowledge bases such as Visual Genome power numerous applications in computer vision, like visual question answering and captioning, but suffer from sparse, incomplete relationships. All scene

Read MoreRead More

Visual Relationships as Functions: Enabling Few-Shot Scene Graph Prediction

Authors Apoorva Dornadula, Austin Narcomey, Ranjay Krishna, Michael Bernstein, Li Fei-Fei We introduce a scene graph approach that formulates predicates as learned functions, which result in an embedding space for objects that is effective for few-shot. Our formulation treats predicates as learned semantic and spatial functions, which are trained within a graph convolution network. First,

Read MoreRead More

Office Hours Announced for CJS Students

The Brown Institute is pleased to announce appointment-based office hours for students needing help in all things digital. This includes (but is not limited to) students seeking assistance with data and statistics, visualization, mapping, natural language processing, web products, immersive media, and the Adobe Creative Suite. Please book office hours at brwn.co/office-hours or by using

Read More

AI-based Request Augmentation to Increase Crowdsourcing Participation

Authors Junwon Park, Ranjay Krishna, Pranav Khadpe, Li Fei-Fei, Michael Bernstein Abstract To support the massive data requirements of modern supervised machine learning (ML) algorithms, crowdsourcing systems match volunteer contributors to appropriate tasks. Such systems learn “what” types of tasks contributors are interested to complete. In this paper, instead of focusing on “what” to ask,

Read MoreRead More

Stanford Welcomes Washington Post’s Director of Engineering

Stanford welcomed Jeremy Bowers, Director of Engineering, at The Washington Post. Bowers and his team are ramping up for the 2020 election, focusing on political data projects including election restyles, congressional votes and campaign finance. Bowers spoke on October 15 to an interdisciplinary group of Stanford students (compute science, engineering and business, among others). He

Read More