0.25
0.5
0.75
1.25
1.5
1.75
2
Knowledge Guided Attention and Inference for Describing Images Containing Unseen Objects
Published on 2018-07-10651 Views
Images on the Web encapsulate diverse knowledge about varied abstract concepts. They cannot be sufficiently described with models learned from image-caption pairs that mention only a small number of v
Related categories
Presentation
Knowledge Guided Attention and Inference for Describing Images Containing Unseen Objects - 100:00
Knowledge Guided Attention and Inference for Describing Images Containing Unseen Objects - 200:33
Knowledge Guided Attention and Inference for Describing Images Containing Unseen Objects - 301:03
Knowledge Guided Attention and Inference for Describing Images Containing Unseen Objects - 402:31
Knowledge guided attention and inference - 103:16
Visual Object Detection03:31
Description Generation for Images04:22
Challenge - Missing Captions for Images05:16
Related Work06:01
Missing in Related Work06:25
Knowledge guided attention and inference - 207:17
Our Contributions07:22
Knowledge-Guided Assistance Caption Generation (KGA-CGM)07:43
External Sematic Attention10:14
TSV Layer11:24
Inference – Generating unseen objects12:58
Evaluation14:40
Evaluation Setup14:55
Qualitative Results - 116:18
Qualitative Results- 217:21
Qualitative Results - 318:06
Scaling it by an order of magnitude18:37
Quantitative Analyse: Out-of-domain Objektbeschreibung - 119:33
Quantitative Analyse: Out-of-domain Objektbeschreibung - 220:11
Reserences and questions20:31