Visual Annotations
Visual Annotation is a visual encoding technique that augments a visualization with supplementary graphical or textual elements to highlight, explain, or contextualize specific data features, distinguished from other annotation forms by its integration within the visual representation space.
531
Publications
50.6K
Citations
1.9K
Authors
546
Institutions
Publications per year
2017–2026
279
Parents
531
DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation
Di Lin, Jifeng Dai, Jiaya Jia et al. · 2016 · 1.1K citations
1–5 of 531
1.9K
8
h-index
6
h-index
6
h-index
6
h-index
6
h-index
1–5 of 1.9K
546
Stanford University
Stanford, United States
11 h-index
39 publications
44.2K citations
University of Michigan
Ann Arbor, United States
11 h-index
26 publications
2.4K citations
Carnegie Mellon University
Pittsburgh, United States
11 h-index
23 publications
1.1K citations
Chinese University of Hong Kong
Hong Kong, Hong Kong
10 h-index
32 publications
12.4K citations
Google (United States)
Mountain View, United States
10 h-index
29 publications
13.4K citations

Chinese Academy of Sciences
Beijing, China
10 h-index
24 publications
1.1K citations
Tsinghua University
Beijing, China
10 h-index
23 publications
1.2K citations

Peking University
Beijing, China
9 h-index
31 publications
1.9K citations
1–8 of 546
Repository · 48 publications · 3.2K citations
25
h-index
TPAMI · 11 publications · 2.5K citations
11
h-index
7 publications · 440 citations
7
h-index
7 publications · 8.4K citations
7
h-index
TVCG · 5 publications · 279 citations
5
h-index
1–5