Take a Prior from Other Tasks for Severe Blur Removal

Abstract

Recovering clear structures from severely blurry inputs is a challenging problem due to the large movements between the camera and the scene. Although some works apply segmentation maps on human face images for deblurring, they cannot handle natural scenes because objects and degradation are more complex, and inaccurate segmentation maps lead to a loss of details. For general scene deblurring, the feature space of the blurry image and corresponding sharp image under the high-level vision task is closer, which inspires us to rely on other tasks (e.g. classification) to learn a comprehensive prior in severe blur removal cases. We propose a cross-level feature learning strategy based on knowledge distillation to learn the priors, which include global contexts and sharp local structures for recovering potential details.In addition, we propose a semantic prior embedding layer with multi-level aggregation and semantic attention transformation to integrate the priors effectively. We introduce the proposed priors to various models, including the UNet and other mainstream deblurring baselines, leading to better per-

AFE-YOLOv8: A Novel Object Detection Model for Unmanned Aerial Vehicle Scenes with Adaptive Feature Enhancement

“…Although shallow feature maps contain detailed feature information about small objects, they also include unnecessary background features, leading to significant redundant computation during the detection process and a slower inference speed of the model. Wang et al [35] and Zhuang et al [21] proposed that deep feature maps contain more semantic information than shallow feature maps. However, deep feature maps have fewer pixels and larger receptive fields, which makes them less suitable for directly detecting small objects in the deep feature layer.…”