Aiming at the problem of leakage and false detection in aerial image detection due to complex background, variable size and insufficient feature extraction, this paper improves YOLOv8n and designs a model named DyFusion-YOLO. It optimises the C2f module by introducing dynamic snake convolution, which can adaptively adjust the convolution kernel morphology and size, and enhance object feature capture capability. In addition, the original spatial pyramid pooling module is replaced with a new module that combines spatial pyramid pooling and cross-stage partial connectivity mechanism to enhance the ability of multi-scale feature fusion and contextual information capture. To address the loss of semantic information due to target scale inconsistency, the study also adds a new small object detection layer to enhance the fusion of deep and shallow semantic information. Meanwhile, a global attention mechanism is introduced at a key position in the neck of the network to make the model focus more on key channels, thus reducing the interference of complex background noise on small object detection. Finally, the model adopts a new loss function, which makes the training pay more attention to high-quality samples and localization accuracy, and improves the detection accuracy. The final results show that the mAP50 of DyFusion-YOLO on VisDrone2019 is improved by 7.0% compared with YOLOv8n, and the computation speed is significantly faster, which is more suitable for aerial object detection tasks.

Scribble-supervised semantic segmentation is an appealing weakly supervised technique with low labeling cost. Existing approaches mainly consider diffusing the labeled region of scribble by low-level feature similarity to narrow the supervision gap between scribble labels and mask labels. In this study, we observe an annotation bias between scribble and object mask, i.e., label workers tend to scribble on the spacious region instead of corners. This label preference makes the model learn well on those frequently labeled regions but poor on rarely labeled pixels. Therefore, we propose BLPSeg to balance the label preference for complete segmentation. Specifically, the BLPSeg first predicts an annotation probability map to evaluate the rarity of labels on each image, then utilizes a novel BLP loss to balance the model training by up-weighting those rare annotations. Additionally, to further alleviate the impact of label preference, we design a local aggregation module (LAM) to propagate supervision from labeled to unlabeled regions in gradient backpropagation. We conduct extensive experiments to illustrate the effectiveness of our BLPSeg. Our single-stage method even outperforms other advanced multi-stage methods and achieves state-of-the-art performance.

Image-level weakly supervised semantic segmentation is a challenging problem that has been deeply studied in recent years. Most of advanced solutions exploit class activation map (CAM). However, CAMs can hardly serve as the object mask due to the gap between full and weak supervisions. In this paper, we propose a self-supervised equivariant attention mechanism (SEAM) to discover additional supervision and narrow the gap. Our method is based on the observation that equivariance is an implicit constraint in fully supervised semantic segmentation, whose pixel-level labels take the same spatial transformation as the input images during data augmentation. However, this constraint is lost on the CAMs trained by image-level supervision. Therefore, we propose consistency regularization on predicted CAMs from various transformed images to provide self-supervision for network learning. Moreover, we propose a pixel correlation module (PCM), which exploits context appearance information and refines the prediction of current pixel by its similar neighbors, leading to further improvement on CAMs consistency. Extensive experiments on PASCAL VOC 2012 dataset demonstrate our method outperforms state-of-the-art methods using the same level of supervision. The code is released online.