MOMA: Multi-Object Multi-Actor Activity Parsing

no code implementations • NeurIPS 2021 • Zelun Luo, Wanze Xie, Siddharth Kapoor, Yiyun Liang, Michael Cooper, Juan Carlos Niebles, Ehsan Adeli, Fei-Fei Li

This paper introduces Activity Parsing as the overarching task of temporal segmentation and classification of activities, sub-activities, atomic actions, along with an instance-level understanding of actors, objects, and their relationships in videos.

Object

OpenTag: Open Attribute Value Extraction from Product Profiles [Deep Learning, Active Learning, Named Entity Recognition]

2 code implementations • 1 Jun 2018 • Guineng Zheng, Subhabrata Mukherjee, Xin Luna Dong, Fei-Fei Li

We study this problem in the context of product catalogs that often have missing values for many attributes of interest.

Active Learning Attribute +4

DDRprog: A CLEVR Differentiable Dynamic Reasoning Programmer

no code implementations • ICLR 2018 • Joseph Suarez, Justin Johnson, Fei-Fei Li

We present a novel Dynamic Differentiable Reasoning (DDR) framework for jointly learning branching programs and the functions composing them; this resolves a significant nondifferentiability inhibiting recent dynamic architectures.

Question Answering Visual Question Answering

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations

1 code implementation • 23 Feb 2016 • Ranjay Krishna, Yuke Zhu, Oliver Groth, Justin Johnson, Kenji Hata, Joshua Kravitz, Stephanie Chen, Yannis Kalantidis, Li-Jia Li, David A. Shamma, Michael S. Bernstein, Fei-Fei Li

Despite progress in perceptual tasks such as image classification, computers still perform poorly on cognitive tasks such as image description and question answering.

Image Classification Question Answering

Love Thy Neighbors: Image Annotation by Exploiting Image Metadata

no code implementations • ICCV 2015 • Justin Johnson, Lamberto Ballan, Fei-Fei Li

Some images that are difficult to recognize on their own may become more clear in the context of a neighborhood of related images with similar social-network metadata.