Authors:
@irit_chelly@shahaffind@ShiraIfergane@OrenFreifeld
We introduce DiTAC, a trainable highly-expressive activation function based on efficient diffeomorphic transformations. DiTAC enhances model expressiveness, using a negligible number of trainable parameters.
Thrilled to share that I will be presenting our paper at #ECCV2024 this week! ๐
Come visit me during my two poster sessions:
- WiCV workshop, today at the Panorama Lounge from 14:00-18:00 pm.
- ECCV poster session on October 3rd from 10:30-12:30 am, poster # 29.
Our ECCV2024 paper, Trainable Highly-expressive Activation Functions, is now available: https://t.co/Zu521BWRgO
@OrenFreifeld@shahaffind Shira Ifergane
We introduce DiTAC, a trainable highly-expressive activation function based on an efficient diffeomorphic transformation.
The 2nd of our 3 #ECCV2024 papers, Trainable Highly-expressive Activation Functions, is now available: https://t.co/5aQhUzwhCq
@shahaffind@irit_chelly Shira Ifergane #deeplearning#computervision
The proposed function, DiTAC, is based on efficient diffeomorphisms (called CPAB).
I'm happy to share that our paper was accepted to #ECCV2024.
In it, we show how wavelet transforms (WT) can be used to achieve CNNs with unprecedented receptive fields and an increased response to low frequencies in the input.
The paper is available at https://t.co/SXsjBMrpmB