computer-vision
Few-Shot Transfer Learning for Saliency Prediction
3260 papers • 126 benchmarks • 313 datasets
Saliency prediction aims to predict important locations in a visual scene. It is a per-pixel regression task with predicted values ranging from 0 to 1. Benefiting from deep learning research and large-scale datasets, saliency prediction has achieved significant success in the past decade. However, it still remains challenging to predict saliency maps on images in new domains that lack sufficient data for data-hungry models.
(Image credit: Papersgraph)
Benchmarks
These leaderboards are used to track progress in few-shot-transfer-learning-for-saliency-prediction
Trend
Dataset
Best Model
Actions
SALICON->WebpageSaliency - 1-shot
SALICON->WebpageSaliency - 1-shot
SALICON->WebpageSaliency - 5-shot
SALICON->WebpageSaliency - 5-shot
Libraries
iUse these libraries to find few-shot-transfer-learning-for-saliency-prediction models and implementations
Datasets
Subtasks
Most implemented papers
n-Reference Transfer Learning for Saliency Prediction
Mohan Kankanhalli, Qi Zhao, Yongkang Wong, Yan Luo•Wed Jul 08 2020
A few-shot transfer learning paradigm for saliency prediction is proposed, which enables efficient transfer of knowledge learned from the existing large-scale saliency datasets to a target domain with limited labeled examples.
6
