Issues

• I am just start to learn, I found that the default configuration of Sparse R-CNN had better AP results than Faster R-CNN. However, with the detection results, I noticed that Sparse R-CNN had lower confidence scores for its detections. Why is that the case? im little confused

  PeizeSun/SparseR-CNN#113 · Filter by author ZhangShutongo opened on May 20, 2024

• 请问怎么使用预训练权重，从头开始训练的准确率比较低

  PeizeSun/SparseR-CNN#112 · Filter by author WinnieMs opened on Mar 19, 2024

• How to visualize the activation maps of dynamic head in Sparse R-CNN？

  PeizeSun/SparseR-CNN#111 · Filter by author Nanakl opened on Oct 31, 2023

• Initialization box selection

  PeizeSun/SparseR-CNN#110 · Filter by author YH-2023 opened on Sep 11, 2023

• Some questions about positional encoding, in the Table 10

  PeizeSun/SparseR-CNN#109 · Filter by author AlterRay opened on Aug 12, 2023

• The results were not reproduced on CrowdHuman dataset

  PeizeSun/SparseR-CNN#108 · Filter by author a1b2c3s4d4 opened on Jun 8, 2023
  1

• "Dataset 'CrowdHuman_train' is not registered

  PeizeSun/SparseR-CNN#107 · Filter by author Aliweka2020 opened on Jun 27, 2022
  3

• KeyError: 1, when using only 1 class dataset to evaluation

  PeizeSun/SparseR-CNN#106 · Filter by author YipChakwai opened on May 24, 2022

• AssertionError: Attribute 'thing_classes' in the metadata of 'coco_2017_train' cannot be set to a different value!

  question
  PeizeSun/SparseR-CNN#102 · Filter by author Aruen24 opened on Jan 28, 2022
  2

• How can I use a custom dataset?

  PeizeSun/SparseR-CNN#99 · Filter by author JEO-96 opened on Nov 26, 2021
  2

• 我把从coco数据集换成了自己的数据集，但是出现了错误，我应该修改你的代码中的哪几个部分的代码

  PeizeSun/SparseR-CNN#98 · Filter by author libingDY opened on Nov 11, 2021
  5

• How to install the Sparse R-CNN(it seems amusing)?

  PeizeSun/SparseR-CNN#97 · Filter by author xiaomin0416 opened on Oct 4, 2021
  1

• How to do use test time augmentation?

  PeizeSun/SparseR-CNN#95 · Filter by author bravezzzzzz opened on Sep 6, 2021
  1

• Does anything changes in detectron2？

  PeizeSun/SparseR-CNN#93 · Filter by author kuiran opened on Jul 12, 2021
  1

• Resume started from zero again!

  PeizeSun/SparseR-CNN#91 · Filter by author zobeirraisi opened on Jul 4, 2021
  3

• How the feature re-use is performed

  PeizeSun/SparseR-CNN#90 · Filter by author XuyangBai opened on Jun 9, 2021
  1

• 'assert (boxes1[:, 2:] >= boxes1[:, :2]).all()' happened when training

  PeizeSun/SparseR-CNN#89 · Filter by author Swall0w opened on May 25, 2021
  4

• how to use plot function?

  PeizeSun/SparseR-CNN#87 · Filter by author Yixiang-Wu opened on May 22, 2021
  1

• Question about the Iterative architecture

  PeizeSun/SparseR-CNN#84 · Filter by author GuangxingHan opened on May 10, 2021
  2

• About the loss format

  PeizeSun/SparseR-CNN#83 · Filter by author FL77N opened on Apr 27, 2021
  4

• train w/o pre-trained backbone

  PeizeSun/SparseR-CNN#82 · Filter by author StarsMyDestination opened on Apr 22, 2021
  3

• question about pred_bboxes.detach() in DynamicHead.forward()

  PeizeSun/SparseR-CNN#81 · Filter by author burningisnaive opened on Apr 20, 2021
  2

• the learning rate adjustment when training with one gpu or two gpus

  PeizeSun/SparseR-CNN#80 · Filter by author Young0111 opened on Apr 1, 2021
  5

• Question regarding proposal feature

  PeizeSun/SparseR-CNN#77 · Filter by author abeyang00 opened on Mar 10, 2021
  4

• issues about Mix-Precision training

  PeizeSun/SparseR-CNN#72 · Filter by author hammer422 opened on Mar 1, 2021
  1