Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring

By Maitreya Suin, Kuldeep Purohit, Ambasamudram Narayanan Rajagopalan. Computer Vision and Pattern Recognition, 2020

328 citations recorded by the source corpus.

This paper tackles the problem of motion deblurring of dynamic scenes. Although end-to-end fully convolutional designs have recently advanced the state-of-the-art in non-uniform motion deblurring, their performance-complexity trade-off is still sub-optimal. Existing approaches achieve a large receptive field by increasing the number of generic convolution layers and kernel-size, but this comesat the expense of of the increase in model size and inference speed. In this work, we propose an efficient pixel adaptive and feature attentive design for handling large blur variations across different spatial locations and process each test image adaptively. We also propose an effective content-aware global-local filtering module that significantly improves performance by considering not only global dependencies but also by dynamically exploiting neighboring pixel information. We use a patch-hierarchical attentive architecture composed of the above module that implicitly discovers the spatial variations in the blur present in the input image and in turn, performs local and global modulation of intermediate features. Extensive qualitative and quantitative comparisons with prior art on deblurring benchmarks demonstrate that our design offers significant improvements over the state-of-the-art in accuracy as well as speed.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f488d9f0d90b5983e1430806ef1cb1edea4e58fe

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

Citations in the corpus

Deep Image Deblurring: A Survey (2022, methodology): “Among these, PSNR and SSIM are the most commonly usedmetrics in image restoration tasks (Gao et al. 2019; Kupyn et al. 2018, 2019; Nah et al. 2017; Shen et al. 2019; Suin et al. 2020; Tao et al. 2018; Zhang et al. 2018a, 2020).”
VRT: A Video Restoration Transformer (2022, methodology): “Method DeepDeblur [58] SRN [78] DMPHN [100] SAPHN [72] MPRNet [99] SFE [93] IFI-RNN [59] ESTRNN [109] EDVR [88] TSP [62] PVDNet [70] GSTA [73] VRT (ours)”
Deblurring via Stochastic Refinement (2021, methodology): “For a fair comparison, we follow the same setup used by [11,14,36,50,65,76] and train our model only using the provided training data.”
Pre-Trained Image Processing Transformer (2020, background): “81 DBGAN [86] MTRNN [57] RADN [58] SAPHN [64] BANET [69] MB2D [56] IPT (Ours) IPT (Ours)”
Denoising Diffusion Restoration Models (2022, background): “Many problems in image processing, including super-resolution [31, 17], deblurring [28, 48], inpainting [55], colorization [29, 58], and compressive sensing [1], are instances of linear inverse problems, where the goal is to recover an image from potentially noisy measurements given through a known ”
Stripformer: Strip Transformer for Fast Image Deblurring (2022, methodology): “In addition, Stripformer can run efficiently without using recurrent architectures compared to the [8,9,33].”
Restormer: Efficient Transformer for High-Resolution Image Restoration (2021, neutral): “[71] MPRNet [93] MIMO-UNet+ [14] Restormer”
Deep Generalized Unfolding Networks for Image Restoration (2022, methodology): “We compare our method with several competitive works [50, 61, 80, 90] and the recent best algorithm MPRNet [79].”
MAXIM: Multi-Axis MLP for Image Processing (2022, methodology): “[81] MPRNet [108] HINet [15] MIMO-UNet [19] MAXIM-3S”