The S-Hock dataset: A new benchmark for spectator crowd analysis

Setti, Francesco; Conigliaro, Davide; Rota, Paolo; Bassetti, Chiara; Conci, Nicola; Sebe, Nicu; Cristani, Marco

doi:10.1016/j.cviu.2017.01.003

Cited by 9 publications

(2 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section we provide both qualitative and quantitative results to validate our proposed methodology. For the evaluation we adopted the S-Hock dataset [16], a publicly available dataset composed by 6 ice-hockey games recorded during the Winter Universiade held in Trentino (Italy) in 2013. This dataset, besides a set of short videos heavily annotated on low level features (e.g.…”

Section: Methodsmentioning

confidence: 99%

Indirect Match Highlights Detection with Deep Convolutional Neural Networks

Godi

Rota

Setti

2017

New Trends in Image Analysis and Processing – ICIAP 2017

Self Cite

View full text Add to dashboard Cite

Abstract. Highlights in a sport video are usually referred as actions that stimulate excitement or attract attention of the audience. A big effort is spent in designing techniques which find automatically highlights, in order to automatize the otherwise manual editing process. Most of the state-of-the-art approaches try to solve the problem by training a classifier using the information extracted on the tv-like framing of players playing on the game pitch, learning to detect game actions which are labeled by human observers according to their perception of highlight. Obviously, this is a long and expensive work. In this paper, we reverse the paradigm: instead of looking at the gameplay, inferring what could be exciting for the audience, we directly analyze the audience behavior, which we assume is triggered by events happening during the game. We apply deep 3D Convolutional Neural Network (3D-CNN) to extract visual features from cropped video recordings of the supporters that are attending the event. Outputs of the crops belonging to the same frame are then accumulated to produce a value indicating the Highlight Likelihood (HL) which is then used to discriminate between positive (i.e. when a highlight occurs) and negative samples (i.e. standard play or time-outs). Experimental results on a public dataset of ice-hockey matches demonstrate the effectiveness of our method and promote further research in this new exciting direction.

show abstract

Section: Methodsmentioning

confidence: 99%

Indirect Match Highlights Detection with Deep Convolutional Neural Networks

Godi

Rota

Setti

2017

New Trends in Image Analysis and Processing – ICIAP 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Rabiee's dataset [31] provides some emotional-rich situations such as panic and fight, although in a staged way. Finally, the S-hock dataset [32] focuses on the behavior of spectator crowds with rich annotations at the individual level, enabling the addition of further affective annotations although restricted to this type of crowds.…”

Section: Crowd Analysis Datasetsmentioning

confidence: 99%

Emotion estimation in crowds: a survey

Urizar

Barakova

Regazzoni

et al. 2017

8th International Conference of Pattern Recognition Systems (ICPRS 2017)

View full text Add to dashboard Cite

Emotions play an important role in human behavior, even more so in large congregations of people where emotional states are prompt to be contaged and amplified. This work presents a qualitative systematic review of the literature concerning the estimation of emotions and affects in real-life crowded environments, covering the aspects of methods and datasets. The academic search engine Scopus was inquired and the search was limited to publications in the English language addressing any of the aforementioned aspects. The aim of this contribution is to highlight advances, limitation and trends in addressing the estimation of emotions in crowds.

show abstract