ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases.
Stéphane d'AscoliHugo TouvronMatthew L. LeavittAri S. MorcosGiulio BiroliLevent SagunPublished in: ICML (2021)
Keyphrases
- computer vision
- vision system
- machine learning
- inductive learning
- real time
- neural network
- active vision
- image processing
- bayesian networks
- visual perception
- network architecture
- object recognition
- inductive reasoning
- rule learning
- human vision
- first order logic
- prior knowledge
- multi agent
- image sequences
- learning algorithm
- data sets