wjn922/ReferFormer

Issues

• None

  #56 opened 3 months ago by Shuaicong97
  0

• WHY is there always NotImplementedError: Cuda is not availabel Error!

  #54 opened 7 months ago by Shuaicong97
  0

• unable to submit results of ref-youtube-vos to the server?

  #41 opened 2 years ago by SleepEarlyLiveLong
  3

• CUDA error: invalid device ordinal

  #55 opened 8 months ago by liuleiBUAA
  1

• How to improve the results of inference/training

  #49 opened 9 months ago by Shuaicong97
  0

• Hello, pre-trained model link 404

  #51 opened 9 months ago by quillbit8
  6

• 0-5 represents the target?

  #53 opened 9 months ago by longmalongma
  2

• can you provide code to generate demo?

  #46 opened a year ago by toyottttttt
  1

• The performance of Video Swin Transformer Base as a backbone on the Ref-Youtube-VOS dataset without pretraining on RefCOCO

  #50 opened 10 months ago by buxiangzhiren
  2

• VisionLanguageFusionModule

  #52 opened 10 months ago by lzl2040
  1

• Is the CUDA operator compilation necessary?

  #48 opened a year ago by Starboy-at-earth
  2

• links

  #45 opened a year ago by ProjectDisR
  1

• None

  #44 opened 2 years ago by yahooo-m
  0

• when i run inference_rvos, i got an error.

  #43 opened 2 years ago by yahooo-m
  0

• What does "Pretrain" mean in the released checkpoints?

  #42 opened 2 years ago by bo-miao
  0

• can you provide log files of pretraining on refcoco/g/+ dataset？

  #40 opened 2 years ago by SleepEarlyLiveLong
  0

• Evaluation issue on A2D-Sentences

  #39 opened 2 years ago by wudongming97
  1

• Performance of joint train with Video-Swin-B

  #35 opened 2 years ago by lszw11
  1

• Vision Language Early fusion

  #37 opened 2 years ago by nero1342
  1

• [Error about finetuning on Ref-Youtube-VOS] No "dataset_name" key in the "targets"

  #38 opened 2 years ago by YOONSEOKHEO
  1

• Frame-based training

  #36 opened 2 years ago by Jay-IPL
  0

• Joint Training Settings

  #34 opened 2 years ago by lszw11
  2

• Finetuning on ref-davis17?

  #33 opened 2 years ago by Jay-IPL
  5

• For A2D-Sentences, did you only use COCO for pre-training or use COCO+YTV?

  #32 opened 2 years ago by bo-miao
  2

• size mismatch when load state_dict from checkpoint

  #31 opened 2 years ago by ja8143912
  0

• question about ref-davis evaluation

  #30 opened 2 years ago by joellliu
  1

• The model with Video-Swin-S backbone trained on Ref-YouTube-VOS dataset is missing

  #29 opened 2 years ago by levon-khachatryan
  2

• ImageNet pre-trained checkpoint for Swin-L

  #28 opened 2 years ago by pvoigtlaender
  0

• RefCOCO/+/g json files

  #25 opened 2 years ago by nankepan
  8

• Mismatch between implementation and conceptual explanation

  #27 opened 2 years ago by basavaraj-hampiholi
  1

• Cannot compile cuda version of MSDeformAttnFunction.

  #26 opened 2 years ago by subramanya1997
  2

• High variance of the reproduced results

  #24 opened 2 years ago by anlimo1510
  0

• how do you get pretrained model

  #19 opened 3 years ago by yuyang16101066
  5

• About sentence feature

  #23 opened 3 years ago by npmhung
  1

• About HMDB51 Dataset

  #17 opened 3 years ago by sutiankang
  2

• Training on DAVIS dataset

  #22 opened 3 years ago by AngelTang190
  1

• No train.zip in ref-youtube-vos dataset

  #21 opened 3 years ago by zhuyan129
  2

• No meta.json in ref-youtube-vos dataset

  #20 opened 3 years ago by zhuyan129
  4

• problem about image pretrain when evaluation

  #15 opened 3 years ago by Bo396543018
  0

• Training time for pre-training phase and main-training phase?

  #18 opened 3 years ago by nero1342
  0

• Initialized weight for video-swin-tiny pretraining

  #16 opened 3 years ago by lxa9867
  1

• CPU Memory increasing when training

  #14 opened 3 years ago by YRlin-12
  2

• joint training hyper parameters

  #7 opened 3 years ago by lxa9867
  9

• How to align the categories of image data and video data during joint training

  #13 opened 3 years ago by Zhiweihhh
  1

• How to use CFBI as post-process

  #12 opened 3 years ago by Zhiweihhh
  1

• Issue w.r.t pretraining models

  #11 opened 3 years ago by youthHan
  3

• Question of Pretraining on RefCOCO/+/g train/val.json

  #10 opened 3 years ago by YRlin-12
  2

• Reproducing training, recommended hardware setup

  #9 opened 3 years ago by pvoigtlaender
  2

• Where to get the test meta file?

  #8 opened 3 years ago by pvoigtlaender
  2

• youtube_vos: No test meta_expressions.json

  #6 opened 3 years ago by Ahnsun
  3