dbolya/tide

Issues

• bbox AP not being calculated

  #55 opened 3 months ago by Changomango0903
  1

• Can I use TIDE on my own dataset - given coco-formatted Ground-Truth and predicted Bounding Boxes?

  #51 opened 2 years ago by codingbutstillalive
  2

• The sum of mAP is not 100

  #53 opened a year ago by duoxiangqinzuo
  3

• Custom dataset evaluation

  #34 opened 3 years ago by gkomix88
  1

• code of Comparison of Scales

  #41 opened 3 years ago by niecongchong
  1

• Test Custom Trained Yolov7 Model

  #54 opened 9 months ago by parth-patel2023
  0

• Even when using the ground truth values as predicted values, there is error.

  #40 opened 3 years ago by abel71
  1

• My Coco files contain no segmentations - only bounding boxes. Why is this incompatible?

  #52 opened 2 years ago by codingbutstillalive
  1

• Crashes with a DivideByZeroError when there are no detections

  #39 opened 3 years ago by drewm1980
  3

• Use case where no segmentation field is present in a COCO file and where COCO result file contains all information about the dataset

  #49 opened 2 years ago by StepanLebedev
  0

• Is there any sample code how to evaluate the detections at different scales and classes? Is there a sample how to use the Qualifier class?

  #48 opened 2 years ago by LaCandela
  0

• Comparing Across Scale

  #33 opened 3 years ago by stuarteiffert
  4

• "IndexError: list index out of range" when run datasets.COCO()

  #20 opened 4 years ago by TommyZihao
  3

• Keypoint Evaluation

  #47 opened 2 years ago by bjlenzig
  0

• where dataset defines

  #45 opened 2 years ago by unfollow7
  0

• About My dataset

  #44 opened 3 years ago by 377fsq
  1

• Getting bbox AP @ 50: 0.00

  #42 opened 3 years ago by meet-minimalist
  1

• How to handle images with no ground truth box

  #43 opened 3 years ago by MagdalenaBaumgaertl
  0

• Dataset which is not COCO in COCO format

  #9 opened 4 years ago by toddChavezz
  2

• TIDE outputs vs. what pycocotools outputs

  #14 opened 4 years ago by kdk2612
  5

• list input can be bounding box (Nx4) or RLEs ([RLE])

  #38 opened 3 years ago by bowmount
  1

• How to implement TIDE for custom dataset?

  #19 opened 4 years ago by LIKHITA12
  7

• Use of pos_threshold

  #15 opened 4 years ago by vincrichard
  3

• Per class plot

  #18 opened 4 years ago by austinmw
  2

• "score" parameter in Data.add_detection is annotated as "int", should probably be "float"

  #37 opened 3 years ago by drewm1980
  0

• Feature Request: Finish statically typing the API

  #36 opened 3 years ago by drewm1980
  1

• Feature Request: Compress masks as they are added to Data

  #35 opened 3 years ago by drewm1980
  0

• Why the summary of dAps and bbox ap is not 100%?

  #32 opened 3 years ago by johnson-magic
  4

• Mismatch of AP as compared to pycocotools due to mismatch number of categories.

  #28 opened 4 years ago by chunchet-ng
  2

• float division by zero Error in td.plot and td.summarize when all errors are 0 and maps are 0

  #16 opened 3 years ago by djaym7
  3

• What does 'suppress' in the paper mean?

  #30 opened 3 years ago by LiewFeng
  1

• Encountered a problem with PASCAL VOC

  #29 opened 4 years ago by LiewFeng
  5

• Is that right to evaluate performance on PASCAL VOC via COCO metric?

  #24 opened 4 years ago by machengcheng2016
  5

• TIDE output interpretation

  #27 opened 4 years ago by jinmingteo
  2

• How to understand "all oracles together results in 100 AP"？

  #26 opened 4 years ago by long2double
  0

• dAP mean what?

  #22 opened 4 years ago by simplew2011
  1

• Recall Metrics

  #25 opened 4 years ago by michael-camilleri
  0

• What fix_main_errors does ?

  #23 opened 4 years ago by mhtarora39
  0

• How to draw fig.4 in the paper

  #17 opened 4 years ago by yuhuixu1993
  0

• ZeroDivisionError

  #10 opened 4 years ago by dvd42
  3

• OpenImages support would be great

  #4 opened 4 years ago by rodrigob
  2

• Quality of Mask for evaluation

  #12 opened 4 years ago by abhigoku10
  2

• Per class recall

  #11 opened 4 years ago by aorad
  8

• What is the difference between "Miss" in "Main Errors" and "FalseNeg" in "Special Errors"

  #13 opened 4 years ago by Yuxin-CV
  2

• BoxError and ClassError can match with used detections ?

  #7 opened 4 years ago by 0phoff
  2

• Dupe detection seems to be working incorrectly

  #6 opened 4 years ago by kirillfish
  2

• No

  #5 opened 4 years ago by fanq15
  0

• ValueError: negative dimensions are not allowed

  #2 opened 4 years ago by hyperparameters
  0