Revisiting Weakly Supervised Pre-Training of Visual Perception Models

01/20/2022
by   Mannat Singh, et al.
0

Model pre-training is a cornerstone of modern visual recognition systems. Although fully supervised pre-training on datasets like ImageNet is still the de-facto standard, recent studies suggest that large-scale weakly supervised pre-training can outperform fully supervised approaches. This paper revisits weakly-supervised pre-training of models using hashtag supervision with modern versions of residual networks and the largest-ever dataset of images and corresponding hashtags. We study the performance of the resulting models in various transfer-learning settings including zero-shot transfer. We also compare our models with those obtained via large-scale self-supervised learning. We find our weakly-supervised models to be very competitive across all settings, and find they substantially outperform their self-supervised counterparts. We also include an investigation into whether our models learned potentially troubling associations or stereotypes. Overall, our results provide a compelling argument for the use of weakly supervised learning in the development of visual recognition systems. Our models, Supervised Weakly through hashtAGs (SWAG), are available publicly.

READ FULL TEXT

page 16

page 18

research
12/06/2019

ClusterFit: Improving Generalization of Visual Representations

Pre-training convolutional neural networks with weakly-supervised and se...
research
12/23/2021

SLIP: Self-supervision meets Language-Image Pre-training

Recent work has shown that self-supervised pre-training leads to improve...
research
06/09/2023

WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction

Most existing word alignment methods rely on manual alignment datasets o...
research
12/07/2020

Self-Supervision Closes the Gap Between Weak and Strong Supervision in Histology

One of the biggest challenges for applying machine learning to histopath...
research
01/11/2021

Wheelchair Behavior Recognition for Visualizing Sidewalk Accessibility by Deep Neural Networks

This paper introduces our methodology to estimate sidewalk accessibiliti...
research
08/07/2023

Exploring Visual Pre-training for Robot Manipulation: Datasets, Models and Methods

Visual pre-training with large-scale real-world data has made great prog...
research
04/20/2023

Complex Mixer for MedMNIST Classification Decathlon

With the development of the medical image field, researchers seek to dev...

Please sign up or login with your details

Forgot password? Click here to reset