# Motion-Aware Adaptive Pixel Pruning for Efficient Local Motion Deblurring

arXiv: https://arxiv.org/abs/2507.07708 · HTML: https://mlsanity.com/paper/2507.07708/

## Key facts

- No citing papers in the map yet (data through 7 Oct 2026).
- Builds on 22 of its 37 references that are in the map.
- First posted on arXiv 10 Jul 2025.

## Builds on (most cited first, 22 of 22)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring](https://mlsanity.com/paper/1612.02177/) (2016)
- [DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks](https://mlsanity.com/paper/1711.07064/) (2017)
- [Scale-recurrent Network for Deep Image Deblurring](https://mlsanity.com/paper/1802.01770/) (2018)
- [Rethinking Coarse-to-Fine Approach in Single Image Deblurring](https://mlsanity.com/paper/2108.05054/) (2021)
- [Learning a Convolutional Neural Network for Non-uniform Motion Blur Removal](https://mlsanity.com/paper/1503.00593/) (2015)
- [Quadratic video interpolation](https://mlsanity.com/paper/1911.00627/) (2019)
- [Stripformer: Strip Transformer for Fast Image Deblurring](https://mlsanity.com/paper/2204.04627/) (2022)
- [Neural Blind Deconvolution Using Deep Priors](https://mlsanity.com/paper/1908.02197/) (2019)
- [ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic](https://mlsanity.com/paper/2103.04039/) (2021)
- [Intriguing Findings of Frequency Selection for Image Deblurring](https://mlsanity.com/paper/2111.11745/) (2021)
- [AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring](https://mlsanity.com/paper/2406.09135/) (2024)
- [Real-World Deep Local Motion Deblurring](https://mlsanity.com/paper/2204.08179/) (2022)
- [LoFormer: Local Frequency Transformer for Image Deblurring](https://mlsanity.com/paper/2407.16993/) (2024)
- [Blur2Blur: Blur Conversion for Unsupervised Image Deblurring on Unknown Domains](https://mlsanity.com/paper/2403.16205/) (2024)
- [Accelerating Image Super-Resolution Networks with Pixel-Level Classification](https://mlsanity.com/paper/2407.21448/) (2024)
- [Adaptive Window Pruning for Efficient Local Motion Deblurring](https://mlsanity.com/paper/2306.14268/) (2023)
- [Aggregating Nearest Sharp Features via Hybrid Transformers for Video Deblurring](https://mlsanity.com/paper/2309.07054/) (2023)

## Abstract (arXiv)

Local motion blur in digital images originates from the relative motion between dynamic objects and static imaging systems during exposure. Existing deblurring methods face significant challenges in addressing this problem due to their inefficient allocation of computational resources and inadequate handling of spatially varying blur patterns. To overcome these limitations, we first propose a trainable mask predictor that identifies blurred regions in the image. During training, we employ blur masks to exclude sharp regions. For inference optimization, we implement structural reparameterization by converting 3× 3 convolutions to computationally efficient 1× 1 convolutions, enabling pixel-level pruning of sharp areas to reduce computation. Second, we develop an intra-frame motion analyzer that translates relative pixel displacements into motion trajectories, establishing adaptive guidance for region-specific blur restoration. Our method is trained end-to-end using a combination of reconstruction loss, reblur loss, and mask loss guided by annotated blur masks. Extensive experiments demonstrate superior performance over state-of-the-art methods on both local and global blur datasets while reducing FLOPs by 49% compared to SOTA models (e.g., LMD-ViT). The source code is available at https://github.com/shangwei5/M2AENet.