Skip to content
This repository has been archived by the owner on Jul 2, 2021. It is now read-only.

Add COCO instance segmentation evaluation #671

Merged
merged 8 commits into from
Jul 9, 2018

Conversation

knorth55
Copy link
Contributor

@knorth55 knorth55 commented Jul 6, 2018

merge after #665

This PR is based on https://github.com/yuyu2172/coco-evaluation

Copy link
Member

@yuyu2172 yuyu2172 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please add doc link

.. csv-table::
:header: name, shape, dtype, format

:obj:`pred_masks`, ":math:`[(R, H, W)]`", :obj:`float32`, \
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bool

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

format is wrong

":math:`[0, \#fg\_class - 1]`"
:obj:`pred_scores`, ":math:`[(R,)]`", :obj:`float32`, \
--
:obj:`gt_masks`, ":math:`[(R, H, W)]`", :obj:`float32`, \
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ditto


This function evaluates predicted bounding boxes obtained from a dataset
by using average precision for each class.
The code is based on the evaluation code used in MS COCO.
Copy link
Member

@yuyu2172 yuyu2172 Jul 9, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bounding boxes ---> instance segmentation

def eval_instance_segmentation_coco(
pred_masks, pred_labels, pred_scores,
gt_masks, gt_labels, gt_areas=None, gt_crowdeds=None):
"""Evaluate detections based on evaluation code of MS COCO.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

detection --> instance segmentation

:header: key, type, description

ap/iou=0.50:0.95/area=all/max_dets=100, *numpy.ndarray*, \
[#coco_det_eval_1]_
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

change coco_det_eval_1 to coco_inst_segm_eval_1.
The link conflicts with eval_detection_coco.
You will notice the problem after compiling.

@knorth55
Copy link
Contributor Author

knorth55 commented Jul 9, 2018

Thank you for review. I updated.

@yuyu2172 yuyu2172 merged commit 8ed189a into chainer:master Jul 9, 2018
@knorth55 knorth55 deleted the coco-ins-eval branch July 9, 2018 12:53
@yuyu2172 yuyu2172 mentioned this pull request Jul 9, 2018
21 tasks
@yuyu2172 yuyu2172 added this to the 0.11 milestone Sep 24, 2018
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants