2012
DOI: 10.1109/tmm.2011.2168948
|View full text |Cite
|
Sign up to set email alerts
|

Semantic Model Vectors for Complex Video Event Recognition

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
129
0
1

Year Published

2013
2013
2020
2020

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 133 publications
(130 citation statements)
references
References 53 publications
0
129
0
1
Order By: Relevance
“…More importantly, it is found that although state-of-the-art concept detections are far from perfect, they still provide useful clues for event classification [54]. [55] also revealed that this representation outperforms -and is complementary to -other low-level visual descriptors for event modeling. This is also tested in a more comprehensive visual lifelogging experiment in [56].…”
Section: Attribute-based Everyday Activity Recognitionmentioning
confidence: 99%
“…More importantly, it is found that although state-of-the-art concept detections are far from perfect, they still provide useful clues for event classification [54]. [55] also revealed that this representation outperforms -and is complementary to -other low-level visual descriptors for event modeling. This is also tested in a more comprehensive visual lifelogging experiment in [56].…”
Section: Attribute-based Everyday Activity Recognitionmentioning
confidence: 99%
“…Các video sẽ được cắt thành các đoạn (shot) mỗi đoạn có thời lượng là 5 giây, trong mỗi đoạn chúng tôi sẽ lấy mẫu theo tần suất 5 cảnh (keyframe)/ giây được làm dữ liệu đầu vào cho quá trình rút trích đặc trưng tiếp theo. Việc lấy mẫu cũng như thông số về thời gian trong một đoạn được sử dụng theo nghiên cứu nhằm đảm bảo mức cân bằng giữa mặt thời gian và độ chính xác sau khi rút trích đặc trưng [19].…”
Section: Tiền Xử Lý Videounclassified
“…We followed the pipeline proposed in [39]. We decoded the videos by uniformly extracting one frame every 2 s. We then applied all available concept detectors to the extracted frames.…”
Section: Visual Conceptsmentioning
confidence: 99%
“…In the future, it is expected that more annotated datasets will be available, and weakly supervised learning methods will help improve the efficiency of generating them. Event representations based on highlevel concepts have started to appear in the literature [13][14][15][16].…”
Section: Introductionmentioning
confidence: 99%