airsplay/py-bottom-up-attention

Issues

• Feature extraction demo (with attributes) not working ... ValueError: not enough values to unpack (expected 3, got 2)

  #25 opened 3 years ago by ig-perez
  2

• cuda 12 not supported

  #29 opened 5 months ago by VamsiYK
  0

• How can i generate label.tsv?

  #19 opened 4 years ago by brightbsit
  5

• No such file or directory: 'data/genome/1600-400-20/objects_vocab.txt'?where is the data?need I download?

  #26 opened 2 years ago by Catrin-baze
  1

• setuptools.errors

  #28 opened a year ago by 569337083
  0

• How can I transfer the label number to its class name?

  #5 opened 5 years ago by coldtomatojuice
  5

• All generated images donot have 36 bounding boxes

  #21 opened 4 years ago by debjitpaul452
  1

• > Great to know that. I have added the link to your project to the Readme file.

  #27 opened 2 years ago by 184446223
  0

• Load weights from original pretrained model

  #1 opened 5 years ago by adrelino
  11

• 'Non-existent config key: MODEL.PROPOSAL_GENERATOR.HID_CHANNELS'

  #4 opened 5 years ago by yestinl
  6

• ValueError cannot reshape array into shape when loading in generated COCO features

  #24 opened 3 years ago by rubencart
  0

• predict 10~100 bbxes in feature extraction

  #23 opened 3 years ago by RubickH
  0

• Coordinates for feature prediction given boxes

  #22 opened 3 years ago by eleonoragualdoni
  1

• Why are the feature dimensions different?

  #20 opened 4 years ago by chenlin038
  1

• How do I extract the features using an FPN based model?

  #7 opened 5 years ago by kevkid
  8

• Integration with LXMERT

  #6 opened 5 years ago by johntiger1
  5

• How do you train this model?

  #16 opened 4 years ago by ThierryDeruyttere
  4

• WARNING [01/11 17:21:52 d2.config.compat]: Config '/home/lvxinyu/AVVP/Others/py-bottom-up-attention/configs/VG-Detection/faster_rcnn_R_101_C4_caffe.yaml' has no VERSION. Assuming it to be compatible with latest v2. Modifications for VG in RPN (modeling/proposal_generator/rpn.py): Use hidden dim 512 instead fo the same dim as Res4 (1024). Modifications for VG in RoI heads (modeling/roi_heads/roi_heads.py): 1. Change the stride of conv1 and shortcut in Res5.Block1 from 2 to 1. 2. Modifying all conv2 with (padding: 1 --> 2) and (dilation: 1 --> 2). For more details, please check 'https://github.com/peteanderson80/bottom-up-attention/blob/master/models/vg/ResNet-101/faster_rcnn_end2end_final/test.prototxt'.

  #18 opened 4 years ago by XinyuLyu
  3

• Getting weights of attribute head from Caffe model

  #2 opened 5 years ago by zhaohengz
  2

• Config '../configs/VG-Detection/faster_rcnn_R_101_C4_caffe.yaml' has no VERSION. Assuming it to be compatible with latest v2.

  #13 opened 4 years ago by mamingyang112
  3

• What is the JSON file "coco_minival_img_ids.json"?

  #17 opened 4 years ago by liuyyy111
  2

• Results different from the original bottom up attention

  #15 opened 4 years ago by zhangchenghua123
  1

• Getting size mismatch message when loading model

  #14 opened 4 years ago by gsrivas4
  1

• How to run detectron2_mscoco_proposal_maxnms.py with multiple gpus?

  #12 opened 4 years ago by Jack-lx-jiang
  2

• how to make Detectron 2 detect not only VGG class names but also ImageNet combined

  #11 opened 5 years ago by monajalal
  0

• How to detect class names that are not in vgg?

  #10 opened 5 years ago by monajalal
  0

• Converting the output tensor which is a Detectron 2 tensor to a normal PyTorch tensor

  #9 opened 5 years ago by monajalal
  1

• Requires AMD ROCM to work (no hipcc found), merged from here into fb/detectron2 as solution

  #8 opened 5 years ago by faizanahemad
  0

• A plan to reveal a Batch-based RoI feature extractor?

  #3 opened 5 years ago by yanan1989
  3