Residual Feature Distillation Network for Lightweight Image Super-Resolution

By Jie Liu, Jie Tang, Gangshan Wu. ECCV Workshops, 2020

582 citations recorded by the source corpus.

Recent advances in single image super-resolution (SISR) explored the power of convolutional neural network (CNN) to achieve a better performance. Despite the great success of CNN-based methods, it is not easy to apply these methods to edge devices due to the requirement of heavy computation. To solve this problem, various fast and lightweight CNN models have been proposed. The information distillation network is one of the state-of-the-art methods, which adopts the channel splitting operation to extract distilled features. However, it is not clear enough how this operation helps in the design of efficient SISR models. In this paper, we propose the feature distillation connection (FDC) that is functionally equivalent to the channel splitting operation while being more lightweight and flexible. Thanks to FDC, we can rethink the information multi-distillation network (IMDN) and propose a lightweight and accurate SISR model called residual feature distillation network (RFDN). RFDN uses multiple feature distillation connections to learn more discriminative feature representations. We also propose a shallow residual block (SRB) as the main building block of RFDN so that the network can benefit most from residual learning while still being lightweight enough. Extensive experimental results show that the proposed RFDN achieve a better trade-off against the state-of-the-art methods in terms of performance and model complexity. Moreover, we propose an enhanced RFDN (E-RFDN) and won the first place in the AIM 2020 efficient super-resolution challenge. Code will be available at this https URL.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/5263d5ebafa1de9942ee1b5a18e909a8221ac48f

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

In Media Technology

Citations in the corpus

Retinex-inspired Unrolling with Cooperative Prior Architecture Search for Low-light Image Enhancement (2020, methodology): “techniques [17], we define our search space as a distillation cell, which is a directed acyclic graph with five nodes and each node connects to the next and the last nodes (see Fig.”
Interpreting Super-Resolution Networks with Local Attribution Maps (2020, methodology): “ﬁelds, PSNR and DI for some representative SR networks. For computing receptive ﬁelds, we follow the guidance provided by Araujo et al. [8]. [31], RCAN, RRDBNet, CARN, SAN, IMDN [25], RNAN [60], RFDN [34], PAN [62], DRLN, and CSNLN [37]. For networks with different network scales, we collect network”
Efficient Long-Range Attention Network for Image Super-resolution (2022, methodology): “To validate the scalability of ELAN, we further compare the normal version of ELAN with state-of-the-art classic performance-oriented SR models, including EDSR [29], SRFBN [27], RNAN [61], RDN [62], RCAN [31], SAN [10], IGNN [64], HAN [41], NLSA [40] and SwinIR [28].”
Transformer for Single Image Super-Resolution (2021, unknown): “To solve this problem, many recurrent networks and lightweight networks have been proposed, such as DRCN [19], SRRFN [23], IMDN [16], IDN [17], CARN [2],ASSLN [46], MAFFSRN [31], and RFDN [27].”