Real-Time Object Tracking via Meta-Learning: Efficient Model Adaptation and One-Shot Channel Pruning

Ilchae Jung
Ilchae Jung
Kihyun You
Kihyun You

AAAI, pp. 11205-11212, 2020.

Cited by: 3|Views40
EI

Abstract:

We propose a novel meta-learning framework for real-time object tracking with efficient model adaptation and channel pruning. Given an object tracker, our framework learns to fine-tune its model parameters in only a few iterations of gradient-descent during tracking while pruning its network channels using the target ground-truth at the...More

Code:

Data:

Your rating :
0

 

Tags
Comments