Object detection on the drone faces a great diversity of challenges such as small object inference, background clutter and wide viewpoint. In contrast to traditional detection problem in computer vision, object detection in bird-like angle can not be transplanted directly from common-in-use methods due to special object texture in sky‘s view. However, due to the lack of a comprehensive data set, the number of algorithms that focus on object detection using data captured by drones is limited. So the VisDrone team gathered a massive data set and organized Vision Meets Drones: A Challenge (VisDrone2021) in conjunction with the IEEE International Conference on Computer Vision (ICCV 2021) to advance the field. The collected dataset is the same as the previous dataset object detection challenge. Specifically, the team needed to predict the bounding boxes of the objects of ten predefined classes. We received results from a number of teams using different approaches, and this article describes the 8 team’s approach. We conducted a detailed analysis of the assessment results and summarized the challenges. More information can be found at: http://www.aiskyeye.com/.