SPAN: Spatial Pyramid Attention Network forImage Manipulation Localization

By Xuefeng Hu, Zhihan Zhang, Zhenye Jiang, Syomantak Chaudhuri, Zhenheng Yang, Ramkant Nevatia. European Conference on Computer Vision, 2020

308 citations recorded by the source corpus.

We present a novel framework, Spatial Pyramid Attention Network (SPAN) for detection and localization of multiple types of image manipulations. The proposed architecture efficiently and effectively models the relationship between image patches at multiple scales by constructing a pyramid of local self-attention blocks. The design includes a novel position projection to encode the spatial positions of the patches. SPAN is trained on a generic, synthetic dataset but can also be fine tuned for specific datasets; The proposed method shows significant gains in performance on standard datasets over previous state-of-the-art methods.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/d2b9d6455bd97c8ed284cd6a638a9015cbd1f9f3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Fahad Shahbaz Khan
2. Saeed Anwar
3. Ajmal Saeed Mian
4. Ming-Hsuan Yang
5. Xiaoming Liu
6. Luc Van Gool
7. Ibrahim Radwan
8. Mohammed Hassanin

In Control and Systems Engineering

1. Ibrahim Radwan
2. Mohammed Hassanin
3. Ajmal Saeed Mian
4. Fahad Shahbaz Khan
5. Saeed Anwar

Citations in the corpus

• PSCC-Net: Progressive Spatio-Channel Correlation Network for Image Manipulation Detection and Localization (2021, methodology): “To finetune PSCC-Net, we follow the same training/testing split on Coverage, CASIA, and NIST16 as in [18],”
• Visual Attention Methods in Deep Learning: An In-Depth Survey (2022, methodology): “Spatial Pyramid Attention Network abbreviated as SPAN [79], was proposed for localizing multiple types of image manipulations.”