Yiru SHEN

How to detect small object in a big picture?

4

### Search before asking - [x] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question Hello. Right now, I have trained my own model,...

结果发生越界问题

4

和其他人员的项目联调时发生了， 越界的错误。m_output_objects_host + bi * (m_param.topK * m_output_objects_width + 1))[0]是个十分大的数，导致int后为-46，用size_t接住查看为 18446744073709551570。但是单独执行我的项目时，无越界状况发生，这种情况下如何判断

### Search before asking - [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question Hello everyone, right now I want to realize an...

双线性插值和数学理论公式差别大

2

我保存了c++和py在双线性插值后的图片，发现预测图片实际有所偏差  ，目标点（122，147）根据数学公式算出核心点应该是（121.30397727272727，138.20738636363635），而在kernel中  w我输出了结果，  在填充方向y上有较大的误差，对于x方向的误差为scale在python和c++的精度不同，在pose模型上没有太大问题，但是在我写的obb的部署中发现预测结果有较大误差

Instance segmentation result at pixel level

7

### Search before asking - [X] I have searched the Ultralytics YOLO [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question Hello, I have trained a YOLOv8-seg model on...