T
Trevor Darrell
Researcher at University of California, Berkeley
Publications - 734
Citations - 222973
Trevor Darrell is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 148, co-authored 678 publications receiving 181113 citations. Previous affiliations of Trevor Darrell include Massachusetts Institute of Technology & Boston University.
Papers
More filters
Proceedings ArticleDOI
Disentangled Action Recognition with Knowledge Bases
TL;DR: This paper proposes the approach: Disentangled Action Recognition with Knowledge-bases (DARK), which leverages the inherent compositionality of actions and has better scalability in the number of objects and verbs.
Proceedings Article
Compositional GAN (Extended Abstract): Learning Image-Conditional Binary Composition
Proceedings ArticleDOI
Learning cross-modal appearance models with application to tracking
John W. Fisher,Trevor Darrell +1 more
TL;DR: An algorithm and experimental results of a human speaker moving in a scene and a method which successfully learns such a model without benefit of hand initialization using only the associated audio signal to "decide" which object to model and track.
Proceedings Article
Predicting with Confidence on Unseen Distributions
TL;DR: In this paper, the authors connect techniques from domain adaptation and predictive uncertainty literature, and predict model accuracy on challenging unseen distributions without access to labeled data, and find that the difference of confidences (DoC) of a classifier's predictions successfully estimates the classifier performance change over a variety of shifts.
Cross-Linked Variational Autoencoders for Generalized Zero-Shot Learning
TL;DR: This work proposes a model where a shared latent space of image features and class embeddings is learned by aligned variational autoencoders, for the purpose of generating latent features to train a softmax classifier.