Skip to search formSkip to main contentSkip to account menu

HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video Clips

It is demonstrated that a text-video embedding trained on this data leads to state-of-the-art results for text-to-video retrieval and action localization on instructional video datasets such as YouCook2 or CrossTask.
View on IEEE (opens in a new tab)

MovieQA: Understanding Stories in Movies through Question-Answering

The MovieQA dataset, which aims to evaluate automatic story comprehension from both video and text, is introduced and existing QA techniques are extended to show that question-answering with such open-ended semantics is hard.
View on IEEE (opens in a new tab)

Semi-supervised Learning with Constraints for Person Identification in Multimedia Data

A unified learning framework is proposed which incorporates labeled and unlabeled data, and constraints between pairs of features in the training, and is applied to train multinomial logistic regression classifiers for multi-class face recognition.
View on IEEE (opens in a new tab)

“Knock! Knock! Who is it?” probabilistic person identification in TV-series

This work model each TV series episode as a Markov Random Field, integrating face recognition, clothing appearance, speaker recognition and contextual constraints in a probabilistic manner, and formulation of the identification task is formulated as an energy minimization problem.
View on IEEE (opens in a new tab)

MovieGraphs: Towards Understanding Human-Centric Situations from Videos

MovieGraphs is the first benchmark to focus on inferred properties of human-centric situations, and opens up an exciting avenue towards socially-intelligent AI agents.
View on IEEE (opens in a new tab)

Total Cluster: A person agnostic clustering method for broadcast videos

The extent to which faces can be clustered automatically without making an error is explored, and an extension of the clustering method to entire episodes using exemplar SVMs based on the negative training data automatically harvested from the editing structure is proposed.
View on ACM (opens in a new tab)

Recovering the Missing Link: Predicting Class-Attribute Associations for Unsupervised Zero-Shot Learning

This work proposes an approach to learn relations that couples class embeddings with their corresponding attributes, given only the name of an unseen class, which outperforms state-of the-art methods in both predicting class-attribute associations and unsupervised ZSL by a large margin.
View on IEEE (opens in a new tab)

Situation Recognition with Graph Neural Networks

A model based on Graph Neural Networks is proposed that allows us to efficiently capture joint dependencies between roles using neural networks defined on a graph and significantly outperforms existing work, as well as multiple baselines.
View on IEEE (opens in a new tab)

Video Face Clustering With Unknown Number of Clusters

Ball Cluster Learning (BCL) is proposed, a supervised approach to carve the embedding space into balls of equal size, one for each cluster, and the learned ball radius is easily translated to a stopping criterion for iterative merging algorithms.
View on IEEE (opens in a new tab)

Think Global, Act Local: Dual-scale Graph Transformer for Vision-and-Language Navigation

This work proposes a dual-scale graph transformer (DUET) for joint long-term action planning and fine-grained cross-modal understanding, and builds a topological map on-the-fly to enable efficient exploration in global action space.
View on IEEE (opens in a new tab)
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)