Search Results for author:
Found 9 papers, 3 papers with code
Medical supervised masked autoencoders: Crafting a better masking strategy and efficient fine-tuning schedule for medical image classification
During the fine-tuning phase, MSMAE is also driven by attention to the accurate masking of medical images.
Star-Net: Improving Single Image Desnowing Model With More Efficient Connection and Diverse Feature Interaction
Compared to other severe weather image restoration tasks, single image desnowing is a more challenging task.
POSTER++: A simpler and stronger facial expression recognition network
POSTER achieves the state-of-the-art (SOTA) performance in FER by effectively combining facial landmark and image features through two-stream pyramid cross-fusion design.
Ranked #3 on
Facial Expression Recognition (FER)
on AffectNet
Facial Expression Recognition
Facial Expression Recognition (FER)
Masked autoencoders are effective solution to transformer data-hungry
This results in ViT not performing as well as CNNs on small datasets like medicine and science.
More comprehensive facial inversion for more effective expression recognition
We extensively evaluate ASIT on facial datasets such as FFHQ and CelebA-HQ, showing that our approach achieves state-of-the-art facial inversion performance.
Facial Expression Recognition
Facial Expression Recognition (FER)
+1
PointCMC: Cross-Modal Multi-Scale Correspondences Learning for Point Cloud Understanding
To solve it, we proposed PointCMC, a novel cross-modal method to model multi-scale correspondences across modalities for self-supervised point cloud representation learning.
Token Transformer: Can class token help window-based transformer build better long-range interactions?
The core mechanism of TT is the addition of a Class (CLS) token for summarizing window information in each local window.
Improvements to Self-Supervised Representation Learning for Masked Image Modeling
The MIM paradigm enables the model to learn the main object features of the image by masking the input image and predicting the masked part by the unmasked part.
Weakly-supervised Generative Adversarial Networks for medical image classification
First, we combine with MixMatch to generate pseudo labels for the fake images and unlabeled images to do the classification.
