MorphMLP: A Self-Attention Free, MLP-Like Backbone for Image and Video.
David Junhao ZhangKunchang LiYunpeng ChenYali WangShashwat ChandraYu QiaoLuoqi LiuMike Zheng ShouPublished in: CoRR (2021)
Keyphrases
- input image
- image data
- multiscale
- image analysis
- image features
- image content
- image retrieval
- image frames
- pre trained
- image pixels
- region of interest
- image representation
- video data
- image collections
- video sequences
- feature points
- multimedia
- spatial information
- image regions
- single image
- high resolution
- segmentation method
- image classification
- test images
- visual data
- static images
- image processing
- images and video sequences
- image registration
- low level
- video retrieval
- dynamic scenes
- artificial neural networks
- video images
- video files