Learning visual biases from human imagination
Although the human visual system can recognize many concepts under challengingconditions, it still has some biases. In this paper, we investigate whether wecan extract these biases and transfer them into a machine recognition system.We introduce a novel method that, inspired by well-known tools in h...
Main Authors: | Vondrick, Carl Martin, Pirsiavash, Hamed, Oliva, Aude, Torralba, Antonio |
---|---|
Other Authors: | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
Format: | Article |
Language: | en_US |
Published: |
Neural Information Processing Systems Foundation
2018
|
Online Access: | http://hdl.handle.net/1721.1/113408 https://orcid.org/0000-0001-5676-2387 https://orcid.org/0000-0003-4915-0256 |
Similar Items
-
Anticipating Visual Representations from Unlabeled Video
by: Vondrick, Carl, et al.
Published: (2018) -
Visualizing Object Detection Features
by: Pirsiavash, Hamed, et al.
Published: (2017) -
Assessing the Quality of Actions
by: Pirsiavash, Hamed, et al.
Published: (2014) -
Generating videos with scene dynamics
by: Vondrick, Carl, et al.
Published: (2020) -
Learning Aligned Cross-Modal Representations from Weakly Aligned Data
by: Castrejon, Lluis, et al.
Published: (2017)