Image Classification
cs.CV · 10 papers
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
FID: 1.73Keyu Tian, Yi Jiang +3
We present Visual AutoRegressive modeling (VAR), a new generation paradigm that redefines the autoregressive learning on images as coarse-to-fine "next-scale prediction" or…
DINOv2: Learning Robust Visual Features without Supervision
ImageNet-1k: 86.5%Maxime Oquab, Timothée Darcet +24
The recent breakthroughs in natural language processing for model pretraining on large quantities of data have opened the way for similar foundation models in computer vision.…
Bootstrap Your Own Latent (BYOL)
ImageNet: 74.3%Jean-Bastien Grill, Florian Strub +12
We introduce Bootstrap Your Own Latent (BYOL), a new approach to self-supervised image representation learning. BYOL relies on two neural networks, referred to as online and…
Momentum Contrast (MoCo)
ImageNet-1M: 60.6%Kaiming He, Haoqi Fan +3
We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic…
SimCLR Explained: A Simple Framework for Contrastive Learning of Visual Representations
ImageNet: 76.5%Ting Chen, Simon Kornblith +2
This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms…
Masked Autoencoders Are Scalable Vision Learners
ImageNet-1K: 87.8%Kaiming He, Xinlei Chen +4
This paper shows that masked autoencoders (MAE) are scalable self-supervised learners for computer vision. Our MAE approach is simple: we mask random patches of the input image…
Deep Double Descent: Where Bigger Models and More Data Hurt
CIFAR-10: 15%Preetum Nakkiran, Gal Kaplun +4
We show that a variety of modern deep learning tasks exhibit a "double-descent" phenomenon where, as we increase model size, performance first gets worse and then gets better.…
High-Resolution Image Synthesis with Latent Diffusion Models
FID: 5.11Robin Rombach, Andreas Blattmann +3
By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image…
An Image is Worth 16x16 Words: Vision Transformers
ImageNet-1K: 88.55%Alexey Dosovitskiy, Lucas Beyer +10
While the Transformer architecture has become the de-facto standard for natural language processing tasks, its applications to computer vision remain limited. In vision,…
Deep Residual Learning for Image Recognition
Top-5 Error: 3.57%Kaiming He, Xiangyu Zhang +2
Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used…