Zobrazeno 1 - 10
of 96
pro vyhledávání: '"Khosla, Aditya"'
Autor:
Baxi, Vipul, Lee, George, Duan, Chunzhe, Pandya, Dimple, Cohen, Daniel N., Edwards, Robin, Chang, Han, Li, Jun, Elliott, Hunter, Pokkalla, Harsha, Glass, Benjamin, Agrawal, Nishant, Lahiri, Abhik, Wang, Dayong, Khosla, Aditya, Wapinski, Ilan, Beck, Andrew, Montalto, Michael
Publikováno v:
In Modern Pathology November 2022 35(11):1529-1539
We propose a general framework called Network Dissection for quantifying the interpretability of latent representations of CNNs by evaluating the alignment between individual hidden units and a set of semantic concepts. Given any CNN model, the propo
Externí odkaz:
http://arxiv.org/abs/1704.05796
Following the gaze of people inside videos is an important signal for understanding people and their actions. In this paper, we present an approach for following gaze across views by predicting where a particular person is looking throughout a scene.
Externí odkaz:
http://arxiv.org/abs/1612.03094
The rise of multi-million-item dataset initiatives has enabled data-hungry machine learning algorithms to reach near-human semantic classification at tasks such as object and scene recognition. Here we describe the Places Database, a repository of 10
Externí odkaz:
http://arxiv.org/abs/1610.02055
Autor:
Krafka, Kyle, Khosla, Aditya, Kellnhofer, Petr, Kannan, Harini, Bhandarkar, Suchendra, Matusik, Wojciech, Torralba, Antonio
From scientific research to commercial applications, eye tracking is an important tool across many domains. Despite its range of applications, eye tracking has yet to become a pervasive technology. We believe that we can put the power of eye tracking
Externí odkaz:
http://arxiv.org/abs/1606.05814
The International Symposium on Biomedical Imaging (ISBI) held a grand challenge to evaluate computational systems for the automated detection of metastatic breast cancer in whole slide images of sentinel lymph node biopsies. Our team won both competi
Externí odkaz:
http://arxiv.org/abs/1606.05718
The complex multi-stage architecture of cortical visual pathways provides the neural basis for efficient visual object recognition in humans. However, the stage-wise computations therein remain poorly understood. Here, we compared temporal (magnetoen
Externí odkaz:
http://arxiv.org/abs/1601.02970
In this work, we revisit the global average pooling layer proposed in [13], and shed light on how it explicitly enables the convolutional neural network to have remarkable localization ability despite being trained on image-level labels. While this t
Externí odkaz:
http://arxiv.org/abs/1512.04150
We introduce algorithms to visualize feature spaces used by object detectors. Our method works by inverting a visual feature back to multiple natural images. We found that these visualizations allow us to analyze object detection systems in new ways
Externí odkaz:
http://arxiv.org/abs/1502.05461
With the success of new computational architectures for visual processing, such as convolutional neural networks (CNN) and access to image databases with millions of labeled examples (e.g., ImageNet, Places), the state of the art in computer vision i
Externí odkaz:
http://arxiv.org/abs/1412.6856