YifanXu74/MQ-Det

Issues

• 编码器输出后的融合问题

  #62 opened 2 months ago by dwr-hub
  0

• 关于在odinw评估问题

  #61 opened 2 months ago by dwr-hub
  0

• 关于单图推理的问题

  #60 opened 3 months ago by KKKSQJ
  0

• 我验证的coco的指标，低于原始的GroundingDino

  #57 opened 4 months ago by TinyQi
  0

• 有没有单图推理的demo呀？

  #56 opened 4 months ago by TinyQi
  0

• GroundingDino使用object365进行Modulated Training，精度不增反降

  #59 opened 4 months ago by TinyQi
  0

• 关于LVIS finetuning-free评估结果的复现

  #44 opened 9 months ago by Real-UtopiaNo
  7

• 关于预训练数据集objects365v2

  #48 opened 8 months ago by CrapbagMo
  2

• The weight of mqdet-grounding-dino

  #58 opened 4 months ago by YWW221
  0

• Conduct modulated pretraining报错

  #41 opened 10 months ago by Swaggeraqua
  1

• Similar performance on increasing number of vision queries for Finetuning Free expts on LVIS

  #55 opened 6 months ago by anishmadan23
  1

• Please provide a simple command to evaluate a single image for detection

  #50 opened 7 months ago by nisyad-ms
  2

• 怎么增加评估时的TEST.IMS_PER_BATCH而不报错？

  #54 opened 6 months ago by Mugealways
  2

• vision query的维度含义

  #53 opened 7 months ago by Mugealways
  1

• 关于finetune时的vision_query

  #52 opened 7 months ago by Real-UtopiaNo
  0

• 关于模型推理过程中视觉查询特征与文本特征的融合增强

  #51 opened 7 months ago by Real-UtopiaNo
  0

• 关于Model Zoo里面的MQ-GLIP-T的问题

  #49 opened 7 months ago by loveforever330
  0

• MQ-Det是否可以用于其他大模型？

  #46 opened 8 months ago by sen2023
  2

• 关于config文件的疑问

  #47 opened 8 months ago by Real-UtopiaNo
  2

• Something is wrong with maskrcnn_benchmark

  #45 opened 8 months ago by AlyssaZyt
  2

• 是否尝试过在其他模型上加入GCP？

  #42 opened 9 months ago by Real-UtopiaNo
  1

• 提取的Vision Query在预训练中怎么调用

  #43 opened 9 months ago by Swaggeraqua
  0

• Unable to start training due to the gpu memory

  #35 opened a year ago by charliess123
  4

• GCP中的QKV的维度

  #40 opened 10 months ago by Mugealways
  5

• cuda version problem

  #39 opened 10 months ago by lanhas
  0

• 关于Vision Query的提取

  #38 opened 10 months ago by NROwind
  0

• Download the object365v1 from OpenDataLab

  #37 opened 10 months ago by zkyseu
  0

• 推理速度

  #36 opened a year ago by ai-learner1
  1

• GCP相关代码在哪个位置呢？

  #34 opened a year ago by Mugealways
  2

• Question about Vision Query Extraction

  #28 opened a year ago by Qia98
  2

• The weight of mqdet-groundingdino

  #33 opened a year ago by er-muyue
  0

• Could you improve the expression in README.md, please？

  #32 opened a year ago by lipond
  1

• Finetune on custom dataset with and without vision query

  #30 opened a year ago by HaisongDing
  6

• 关于论文 LVIS 结果的一个问题

  #31 opened a year ago by Cycyes
  7

• finetune in costum datasets

  #27 opened a year ago by Qia98
  2

• why two cross-attention？

  #29 opened a year ago by PAradoxLG
  1

• Calculate the parameters and FLOPs (including GLIP and GroundingDINO)

  #26 opened a year ago by Qia98
  1

• 关于object365_query_5000_sel_tiny_old问题请教

  #25 opened a year ago by wangmedr
  7

• 如何使用 Modulated pre-training后的模型来检测少量图片上的效果？

  #24 opened a year ago by Mugealways
  2

• 检测任意图片上的感兴趣类别区域

  #23 opened a year ago by ai-learner1
  1

• Vision Query Extraction for Custom dataset

  #20 opened a year ago by Qia98
  1

• 请问视觉端Q是在哪输入的

  #22 opened a year ago by wangmedr
  1

• Finetuning-free eval COCO datasets

  #21 opened a year ago by Qia98
  1

• setup error

  #17 opened a year ago by Qia98
  4

• 使用COCO数据集训练后如何评测性能？

  #16 opened a year ago by Mugealways
  4

• eval_odinw测试出错

  #19 opened a year ago by wangmedr
  2

• -c.modulated_deform_conv_forward() 报RuntimeError:Not comiled with GPU support ,但我GPU跟pytorch版本一致

  #18 opened a year ago by chaochao1993
  1

• 请问多显卡单节点怎么训练

  #15 opened a year ago by chaochao1993
  1

• GroundingDINO的训练代码里有DN部分吗？

  #14 opened a year ago by Baboom-l
  3

• 求助

  #12 opened a year ago by daimazhnenana
  4