ImageNet 64 Model

ImageNet-64 models represent a significant area of research focused on improving the robustness and efficiency of image recognition systems, particularly within the context of the ImageNet dataset. Current research emphasizes enhancing model architectures, such as Vision Transformers (ViTs), through techniques like wavelet transforms for improved multi-scale feature extraction and the incorporation of data from other modalities to boost performance. A key focus is understanding and mitigating model failures by analyzing the influence of factors like pose, lighting, and background, leading to the development of datasets annotated with these variations. These advancements aim to create more reliable and accurate image recognition models with broader practical applications.

Papers

January 25, 2024

Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities
Yiyuan Zhang, Xiaohan Ding, Kaixiong Gong, Yixiao Ge, Ying Shan, Xiangyu Yue
Transformer Megatron Decepticons Irrelevant Information Target Modality Modal Fine Tuning Multimodal Goal ImageNet 64 Model

November 3, 2022

ImageNet-X: Understanding Model Mistakes with Factor of Variation Annotations
Badr Youbi Idrissi, Diane Bouchacourt, Randall Balestriero, Ivan Evtimov, Caner Hazirbas, Nicolas Ballas, Pascal Vincent, Michal Drozdzal, David Lopez-Paz, Mark Ibrahim
Vision Model Key Factor ImageNet 1k Recognition Model Model Error ImageNet 64 Model

July 11, 2022

Wave-ViT: Unifying Wavelet and Transformers for Visual Representation Learning
Ting Yao, Yingwei Pan, Yehao Li, Chong-Wah Ngo, Tao Mei
Vision Transformer Transformer Megatron Decepticons Visual Representation Learning Wavelet Transform Self Attention Computation Multi Scale Vision Transformer ImageNet 64 Model

June 1, 2022

Elucidating the Design Space of Diffusion-Based Generative Models
Tero Karras, Miika Aittala, Timo Aila, Samuli Laine
Generative Model Modularity Function Design Space Diffusion Based Generative Model Score Estimation Network ImageNet 64 Model

March 28, 2022

Core Risk Minimization using Salient ImageNet
Sahil Singla, Mazda Moayeri, Soheil Feizi
Supervised ImageNet High Performing 1D ConvNet Risk Minimization ImageNet Class ImageNet 64 Model

ImageNet 64 Model

Papers

Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities

ImageNet-X: Understanding Model Mistakes with Factor of Variation Annotations

Wave-ViT: Unifying Wavelet and Transformers for Visual Representation Learning

Elucidating the Design Space of Diffusion-Based Generative Models

Core Risk Minimization using Salient ImageNet