open-mmlab/mmdetection

Issues

• Box Predictions - Confidence Thresholding during Evaluation

  #12016 opened 2 months ago by laurenzheidrich
  2

• 加载grounding-dino大模型进行推理时，一张图需要30s也太慢了吧

  #11969 opened 3 months ago by jamesbondzhou
  2

• Zero mAP during validation when using ViT backbone in Faster R-CNN

  #12017 opened 2 months ago by chan1031
  5

• how to use Copy-paste in mmdetection3.3.0

  #11994 opened 3 months ago by liruixinxinxin
  1

• keyerror in training mm-grounding-dino

  #11971 opened 3 months ago by xingmimfl
  1

• Issue when trying to train MM_Grounding_DINO

  #12024 opened 2 months ago by saraalemadi
  1

• Grounding-Dino: BERT model frozen or not?

  #12010 opened 2 months ago by laurenzheidrich
  2

• pycocotools and runner.test() scores do not match

  #12009 opened 2 months ago by daniellehot
  1

• How to use 3 coco datasets in my train?

  #11995 opened 3 months ago by Hecoming
  2

• How to convert MMdetection model to PyTorch?

  #12023 opened 2 months ago by nikitaSisikin
  0

• KeyError: 'Adafactor is already registered in optimizer at torch.optim'

  #12008 opened 2 months ago by mobguang
  6

• how does this category information keep up to date during the interface multi-task training run?

  #12011 opened 2 months ago by 1wang11lijian1
  3

• maskrcnn训练自己的数据小目标非常非常差，检测不出来

  #12012 opened 2 months ago by bo-bobo
  1

• Is there any ideas to speed up analyze_results.py?

  #12022 opened 2 months ago by gjustin40
  0

• ImportError: cannot import name 'make_res_layer' from 'mmdet.models.backbones'

  #12020 opened 2 months ago by jiyuwangbupt
  0

• FInetune model outputs only one class at the time

  #12019 opened 2 months ago by SMohammadi89
  0

• Problem about checkpoint auto saving

  #12018 opened 2 months ago by BrokenArrow1404
  0

• load from not working?

  #12001 opened 3 months ago by JNaranjo-Alcazar
  1

• type='RepeatDataset' in faster-rcnn r18 How to close it

  #12015 opened 2 months ago by DaCheng1823
  1

• Modify Config file for Custom Modules

  #12002 opened 3 months ago by Mukil07
  1

• About some issues

  #11966 opened 3 months ago by wap1024
  2

• SwinL weights for Grounding DINO

  #12014 opened 2 months ago by german36-del
  0

• the question of voc_metric.py

  #12013 opened 2 months ago by anzitong
  0

• ModuleNotFoundError: No module named 'mmcv._ext'

  #11983 opened 3 months ago by mbspng
  2

• Get TypeError: __init__() got an unexpected keyword argument 'data_root' while using RepeatDataset

  #12007 opened 2 months ago by Icecream-blue-sky
  0

• Does Co-DETR supports automatic-mixed-precision training?

  #12000 opened 3 months ago by taemmini
  0

• How to get the values of query and key in the attention module?

  #11999 opened 3 months ago by jiugexuan
  0

• error raise when finetuning groundingdino 微调groudingdino train test报错

  #11974 opened 3 months ago by xuans29
  0

• set start_epoch/start_iter

  #11998 opened 3 months ago by YasminZhang
  0

• 真的崩溃了兄弟们

  #11996 opened 3 months ago by Zihan-Liu-westlake
  0

• How to use the Mask2Former model for semantic segmentation?

  #11997 opened 3 months ago by Invincible-predator
  0

• How to use the mmpretrain model in mmdetection?

  #11978 opened 3 months ago by sjtu-cz
  3

• data'num changes

  #11993 opened 3 months ago by Hecoming
  1

• RepeatDataset.__init__() got an unexpected keyword argument 'pipeline'

  #11991 opened 3 months ago by QianYing-LYG
  0

• DINO missing detection

  #11980 opened 3 months ago by auindoeee
  2

• About grounding dino

  #11990 opened 3 months ago by xiang-xiang-zhu
  0

• Can parallel inference be used in dino detection?

  #11989 opened 3 months ago by zhoulin2545210131
  0

• how to modify config.py for mot inference?

  #11988 opened 3 months ago by hoya-cho
  0

• The configuration file of the mmdetection v3 version is out of order

  #11987 opened 3 months ago by liuhongyan123456
  0

• Compatibility Issue with MMCV, PyTorch, and CUDA in Google Colab Object Detection Model

  #11984 opened 3 months ago by faizan628
  2

• AssertionError: The num_classes (3) in SSDHead of SingleStageDetector does not matches the length of classes 80) in CocoDataset

  #11986 opened 3 months ago by psantiago-lsbd
  0

• YOLOX finetuning - COCO dataset - losses at zero - ERROR from `compute_metrics` - the whole dataset is empty

  #11985 opened 3 months ago by LukasHassel
  0

• Security concern

  #11982 opened 3 months ago by psmoros
  0

• training grounding dino with 139 classes

  #11975 opened 3 months ago by sheethalb
  2

• Analysis of the Discrepancy Between Precision/Recall from Confusion Matrix and mAP During Training

  #11977 opened 3 months ago by liruixinxinxin
  1

• about transformer output

  #11979 opened 3 months ago by jiugexuan
  0

• mmdetection is getting more and more rubbish

  #11976 opened 3 months ago by 2308700388
  0

• 2.x's<img_metas>change to which in 3.x?

  #11973 opened 3 months ago by Hecoming
  1

• 'DetDataSample' object has no attribute 'text' when finetune on grounding task

  #11972 opened 3 months ago by xingmimfl
  0

• cascade mask-rcnn + mmdeploy error: [mmdeploy] [error] [instance_segmentation.cpp:78] invalid argument (1)?

  #11967 opened 3 months ago by yangzhiyin
  0