D
Dhruv Mahajan
Researcher at Facebook
Publications - 74
Citations - 5206
Dhruv Mahajan is an academic researcher from Facebook. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 27, co-authored 68 publications receiving 3554 citations. Previous affiliations of Dhruv Mahajan include Microsoft & Columbia University.
Papers
More filters
Book ChapterDOI
Exploring the Limits of Weakly Supervised Pretraining
Dhruv Mahajan,Ross Girshick,Vignesh Ramanathan,Kaiming He,Manohar Paluri,Yixuan Li,Ashwin Bharambe,Laurens van der Maaten +7 more
TL;DR: In this paper, the authors presented a transfer learning approach with large convolutional networks trained to predict hashtags on billions of social media images and reported the highest ImageNet-1k single-crop, top-1 accuracy to date.
Posted Content
Exploring the Limits of Weakly Supervised Pretraining
Dhruv Mahajan,Ross Girshick,Vignesh Ramanathan,Kaiming He,Manohar Paluri,Yixuan Li,Ashwin Bharambe,Laurens van der Maaten +7 more
TL;DR: This paper presents a unique study of transfer learning with large convolutional networks trained to predict hashtags on billions of social media images and shows improvements on several image classification and object detection tasks, and reports the highest ImageNet-1k single-crop, top-1 accuracy to date.
Posted Content
Billion-scale semi-supervised learning for image classification.
TL;DR: This paper proposes a pipeline, based on a teacher/student paradigm, that leverages a large collection of unlabelled images to improve the performance for a given target architecture, like ResNet-50 or ResNext.
Proceedings ArticleDOI
Scaling and Benchmarking Self-Supervised Visual Representation Learning
TL;DR: It is shown that by scaling on various axes (including data size and problem 'hardness'), one can largely match or even exceed the performance of supervised pre-training on a variety of tasks such as object detection, surface normal estimation and visual navigation using reinforcement learning.
Posted Content
Scaling and Benchmarking Self-Supervised Visual Representation Learning.
TL;DR: In this article, the authors show that by scaling on various axes (including data size and problem 'hardness'), self-supervised learning can largely match or even exceed the performance of supervised pre-training on a variety of tasks such as object detection, surface normal estimation (3D) and visual navigation using reinforcement learning.