hkchengrex/Tracking-Anything-with-DEVA

Issues

• mask problem

  #101 opened 6 days ago by HibaDoi
  1

• Cannot segment and track Gundam. Any suggestion? Thanks

  #100 opened 7 days ago by garrisonz
  1

• Floor tracking is poor. Any suggestion?

  #98 opened 21 days ago by garrisonz
  2

• How to Perform Inference on Multiple Videos?

  #99 opened 15 days ago by weilllllls
  1

• RuntimeError: shape '[1, 63, 512, 20, 27]' is invalid for input of size 19353600

  #97 opened 16 days ago by GuanghaoJin
  2

• something about data format

  #85 opened 2 months ago by wkywwds
  29

• Will changing the batch size from 16 to 18 and running with 6 cards result in performance loss?

  #96 opened a month ago by longmalongma
  3

• RuntimeWarning: Trying to segment without any memory

  #90 opened a month ago by Masrur02
  6

• Can deva be extended to video semantic segmentation and video instance segmentation tasks?

  #92 opened a month ago by longmalongma
  4

• reading data is particularly slow

  #94 opened a month ago by longmalongma
  2

• Hello, I found that this method matches black results when there are only two images of this input.。。

  #93 opened a month ago by chenkangjie1123
  1

• Subsampling BURST when evaluating

  #91 opened a month ago by chhchan
  1

• SAM for the training sets of YouTube VOS and Davis 2017

  #89 opened a month ago by longmalongma
  2

• Something about the provided detection on VIPSeg

  #88 opened a month ago by chhchan
  3

• How to get subpart/part/whole-level segmentation like SAM？

  #86 opened 2 months ago by chenkangjie1123
  2

• The mask in example

  #87 opened 2 months ago by wkywwds
  3

• RuntimeWarning: Trying to segment without any memory!

  #84 opened 2 months ago by Masrur02
  7

• extracting mask and bbox

  #83 opened 2 months ago by shani1610
  1

• Just want to confirm!

  #81 opened 2 months ago by JawadTawhidi
  5

• yolo_world with DEVA

  #78 opened 2 months ago by LLH-Harward
  2

• Just want to confirm!

  #82 opened 2 months ago by JawadTawhidi
  1

• U-net with DEVA

  #79 opened 3 months ago by wkywwds
  13

• Parameter Adjustment in Automatic Mode and Objects Reappearance Problem

  #80 opened 2 months ago by dhgras
  2

• TrustImageSeg Inference

  #77 opened 3 months ago by SangbeomLim
  1

• 0-5 represents the target?

  #76 opened 3 months ago by longmalongma
  4

• Didn't you use a mask that was inferred from a large model during semi-youtubevos inference?

  #75 opened 3 months ago by longmalongma
  3

• target tracking count

  #74 opened 3 months ago by YuqiuDeng
  1

• How do I reproduce the results of the rvos report？

  #73 opened 3 months ago by xinbaiw
  10

• how to get the mask path to detections for rvos

  #72 opened 3 months ago by xinbaiw
  2

• Using DEVA with a different mask model causing issues with mask encoding

  #65 opened 3 months ago by vineetparikh
  3

• Wrong segmentation appears after several objects come and go from the scene

  #64 opened 3 months ago by be-hd
  6

• Error in the middle of a video.

  #67 opened 3 months ago by Mountchicken
  2

• valueerror of text prompt

  #70 opened 3 months ago by shangchengPKU
  2

• text prompt

  #69 opened 3 months ago by yangyang117
  1

• Deva out of memory

  #54 opened 3 months ago by zhangzehao123
  3

• High Memory Usage During Initialization

  #68 opened 3 months ago by kaankarakose
  1

• Replace [grounding-dino+sam] with [yolo-world+efficient-sam]

  #66 opened 3 months ago by guoqincode
  1

• Segmentation broken after 20 seconds or when total about more than 30 objects appeared in the video

  #63 opened 4 months ago by huyen-do-helm
  1

• Crop / segment the objects

  #62 opened 4 months ago by tralala87
  1

• Deva on CPU

  #61 opened 4 months ago by riccardo24-7
  1

• Is there a way to use DEVA with Cutie?

  #58 opened 5 months ago by goneflash
  3

• How to inference a video directly

  #59 opened 5 months ago by wuyiqii
  2

• Licence

  #60 opened 5 months ago by makemate
  1

• Pre-trained model

  #56 opened 5 months ago by cwt000297
  1

• None in category_ids

  #57 opened 6 months ago by Bariona
  2

• Not working if there is large variation between consecutive frames??

  #53 opened 6 months ago by JawadTawhidi
  5

• Deva run out of memory with "Including non-PyTorch memory, this process has 17179869184.00 GiB memory in use."

  #55 opened 6 months ago by zhangzehao123
  1

• Is it normal for DEVA's training speed to be much slower than XMem?

  #52 opened 6 months ago by longmalongma
  5

• Why is the training speed of DEVA much slower than XMem?

  #51 opened 7 months ago by longmalongma
  3

• About training details

  #50 opened 7 months ago by csustYyh
  1