Learning from the limited amount of labeled data to the pre-train model has
always been viewed as a challenging task. In this report, an effective and
robust solution, the two-stage training paradigm YOLOv8 detector (TP-YOLOv8),
is designed for the object detection track in VIPriors Challenge 2023. First,
the backbone of YOLOv8 is pre-trained as the encoder using the masked image
modeling technique. Then the detector is fine-tuned with elaborate
augmentations. During the test stage, test-time augmentation (TTA) is used to
enhance each model, and weighted box fusion (WBF) is implemented to further
boost the performance. With the well-designed structure, our approach has
achieved 30.4% average precision from 0.50 to 0.95 on the DelftBikes test set,
ranking 4th on the leaderboard.

这篇报告介绍了一种有效且稳健的解决方案，即 VIPriors Challenge 2023 中物体检测任务的两阶段训练范式 YOLOv8 检测器（TP-YOLOv8），通过有限的标记数据对预训练模型进行学习，利用掩模图像建模技术将 YOLOv8 的主干部分作为编码器进行预训练，再通过精细调整的增强技术对检测器进行微调，在测试阶段利用测试时增强和加权框融合来提高性能，在 DelftBikes 测试集上取得了 0.50 到 0.95 之间 30.4% 的平均精确度，排名第 4。