$n$-Reference Transfer Learning for Saliency Prediction Article Swipe

MAKE IMPACT

Yan Luo , Yongkang Wong , Mohan Kankanhalli , Qi Zhao · YOU? · · 2020 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2007.05104

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. To solve this problem, we propose a few-shot transfer learning paradigm for saliency prediction, which enables efficient transfer of knowledge learned from the existing large-scale saliency datasets to a target domain with limited labeled examples. Specifically, very few target domain examples are used as the reference to train a model with a source domain dataset such that the training process can converge to a local minimum in favor of the target domain. Then, the learned model is further fine-tuned with the reference. The proposed framework is gradient-based and model-agnostic. We conduct comprehensive experiments and ablation study on various source domain and target domain pairs. The results show that the proposed framework achieves a significant performance improvement. The code is publicly available at \url{https://github.com/luoyan407/n-reference}.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2007.05104

PDF

https://arxiv.org/pdf/2007.05104

OA Status

green

Cited By

1

References

51

Related Works

10

OpenAlex ID

https://openalex.org/W3041094932

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W3041094932

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2007.05104

Digital Object Identifier

Title

$n$-Reference Transfer Learning for Saliency Prediction

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2020

Year of publication

Publication date

2020-07-09

Full publication date if available

Authors

Yan Luo, Yongkang Wong, Mohan Kankanhalli, Qi Zhao

List of authors in order

Landing page

https://arxiv.org/abs/2007.05104

Publisher landing page

PDF URL

https://arxiv.org/pdf/2007.05104

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2007.05104

Direct OA link when available

Concepts

Computer science, Domain (mathematical analysis), Artificial intelligence, Transfer of learning, Code (set theory), Process (computing), Source code, Machine learning, Scale (ratio), Labeled data, Pattern recognition (psychology), Data mining, Mathematics, Operating system, Mathematical analysis, Physics, Quantum mechanics, Set (abstract data type), Programming language

Top concepts (fields/topics) attached by OpenAlex

Cited by

1

Total citation count in OpenAlex

Citations by year (recent)

2021: 1

Per-year citation counts (last 5 years)

References (count)

51

Number of works referenced by this work

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload