Joint Super-Resolution and Inverse Tone-Mapping: A Feature Decomposition Aggregation Network and A New Benchmark

TL;DR

This paper introduces a novel feature decomposition network for joint super-resolution and inverse tone-mapping, along with a new large-scale dataset, demonstrating superior performance over existing methods.

Contribution

We propose a lightweight, learnable feature decomposition network and a new benchmark dataset for joint SR-ITM, enhancing versatility and evaluation in this task.

Findings

FDAN outperforms state-of-the-art methods on benchmark datasets.

The new SRITM-4K dataset provides diverse scenarios for robust evaluation.

Our approach is efficient and effective for joint SR-ITM tasks.

Abstract

Joint Super-Resolution and Inverse Tone-Mapping (joint SR-ITM) aims to increase the resolution and dynamic range of low-resolution and standard dynamic range images. Recent networks mainly resort to image decomposition techniques with complex multi-branch architectures. However, the fixed decomposition techniques would largely restricts their power on versatile images. To exploit the potential power of decomposition mechanism, in this paper, we generalize it from the image domain to the broader feature domain. To this end, we propose a lightweight Feature Decomposition Aggregation Network (FDAN). In particular, we design a Feature Decomposition Block (FDB) to achieve learnable separation of detail and base feature maps, and develop a Hierarchical Feature Decomposition Group by cascading FDBs for powerful multi-level feature decomposition. Moreover, to better evaluate the comparison methods, we collect a large-scale dataset for joint SR-ITM, i.e., SRITM-4K, which provides versatile scenarios for robust model training and evaluation. Experimental results on two benchmark datasets demonstrate that our FDAN is efficient and outperforms state-of-the-art methods on joint SR-ITM. The code of our FDAN and the SRITM-4K dataset are available at https://github.com/CS-GangXu/FDAN.