Search results

Custom backbone: Very low mAP (~0) and recall (~0) during training

Status: Open.

#12375 In open-mmlab/mmdetection;

· LinkZY3471 opened on Jun 15, 2025

How can I augment the image while keeping the face intact?

Status: Open.

#12374 In open-mmlab/mmdetection;

· wzr0108 opened on Jun 13, 2025

MMdetection3.0 how to add AP30

Status: Open.

#12373 In open-mmlab/mmdetection;

· GuanyuChen8 opened on Jun 12, 2025

mmpose1.3.2、mmdet3.3.0与mmcv2.2.0不兼容

Status: Open.

#12372 In open-mmlab/mmdetection;

· namofoo opened on Jun 12, 2025

训练过程中显存占满，验证时出现out of memory

Status: Open.

#12370 In open-mmlab/mmdetection;

· SwustLiC opened on Jun 4, 2025

有关被截断目标预测的问题

Status: Open.

#12369 In open-mmlab/mmdetection;

· zhongqiu1245 opened on Jun 1, 2025

多模态glip demo推理报错

Status: Open.

#12368 In open-mmlab/mmdetection;

· Hitheigy opened on May 28, 2025

KeyError: 'FasterRCNN is not in the mmengine::model registry. Please check whether the value of <code>FasterRCNN</code> is correct or it was registered as expected.

reimplementationIssues in model reimplementationIssues in model reimplementation

Status: Open.

#12367 In open-mmlab/mmdetection;

· meixiangzhang opened on May 27, 2025

how to use gradient accumulation?

Status: Open.

#12366 In open-mmlab/mmdetection;

· helloyaworld opened on May 27, 2025

At present, the 50 series graphics card only supports Cuda12.8 and Torch2.8, so how should I choose the corresponding mmdet and mmcv?

Status: Open.

#12365 In open-mmlab/mmdetection;

· kNDSFLJ opened on May 25, 2025

关于copy_paste数据增强方案

Status: Open.

#12362 In open-mmlab/mmdetection;

· RobinHan24 opened on May 17, 2025

why not a tutorial to transform code base mmdet 2.0 to mmdet 3.0??? Lots of incompatible issues!!

Status: Open.

#12361 In open-mmlab/mmdetection;

· kasteric opened on May 17, 2025