Skip to content

40000轮visualization/eval效果不一致 #28

Description

@KevinSONG729

您好,我在复现代码过程中遇到了下述问题:
我使用kitti_mot数据集中的0002场景(所有图片)进行训练,并在40000轮训练结束后,运行了evaluate.py代码:
python train.py --config xxx/kitti_reconstruction.yaml source_path=xxx model_path=xxx start_frame=0 end_frame=232
python evaluate.py --config xxx/kitti_reconstruction.yaml source_path=xxx model_path=xxx start_frame=0 end_frame=232
yaml文件配置保持原始设置。但是发现训练过程中输出的visualization文件夹中的图片以及训练结束后eval文件夹中的图片,和使用evaluation.py输出的图片效果不一致,后者在非天空的区域明显变暗:
visiualization/40000_148.png:
image
训练结束时自动eval的结果187.png:
image
手动运行evaluate.py的结果187.png(40000iter):
image
手动运行evaluate.py的结果187.png(7000iter):
image
发现7000iter时evaluation.py的运行结果比较合理,可是40000iter的evaluation.py输出结果与训练阶段的输出结果出现了明显的不一致。因此想要请问下,出现上述问题是否是哪里的配置不当导致的,如何修复该错误?非常期待您的回复!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions