en
0.25
0.5
0.75
1.25
1.5
1.75
2
Microsoft COCO: Common Objects in Context
Published on Oct 29, 20142874 Views
We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understandi
Related categories
Chapter list
Microsoft COCO00:00
Why a new dataset? - 100:29
Why a new dataset? - 200:53
Why a new dataset? - 301:09
Why a new dataset? - 401:22
Iconic object images01:45
Iconic scene images01:58
Non-iconic images - 102:10
Non-iconic images - 202:43
Non-iconic images - 303:21
Object categories03:42
flickr04:20
“Dog”04:45
“Dog + Car”05:05
Annotation pipeline - 105:40
Annotation pipeline - 205:49
Annotation pipeline - 306:01
Divide and Conquer06:10
1. Category Labeling06:21
2. Instance Spotting07:01
3. Instance Segmentation - 107:29
3. Instance Segmentation - 207:43
3. Instance Segmentation - 307:52
3. Instance Segmentation - 407:57
3. Instance Segmentation - 508:08
Properties08:48
Number of categories vs. number of instances - 108:55
Number of categories vs. number of instances - 209:09
Number of categories vs. number of instances - 309:24
Number of categories vs. number of instances - 409:31
Categories per image09:49
Detection Performance10:25
http://mscoco.org - 110:57
Beyond detection - 111:16
Beyond detection - 211:37
Beyond detection - 311:48
How to use COCO12:11
http://mscoco.org - 212:30
APIs12:47
MS COCO 2014 release12:55
Going forward13:11
Algorithm Evaluation13:14
MS COCO 201513:29
Thank you!13:40