TY - JOUR
T1 - Rethinking Online Action Detection in Untrimmed Videos
T2 - A Novel Online Evaluation Protocol
AU - Baptista Ríos, Marcos
AU - Lopez-Sastre, Roberto J.
AU - Caba Heilbron, Fabian
AU - van Gemert, Jan C.
AU - Acevedo-Rodriguez, F. Javier
AU - Maldonado-Bascon, Saturnino
N1 - green
PY - 2019
Y1 - 2019
N2 - The Online Action Detection (OAD) problem needs to be revisited. Unlike traditional offline action detection approaches, where the evaluation metrics are clear and well established, in the OAD setting we find very few works and no consensus on the evaluation protocols to be used. In this work we propose to rethink the OAD scenario, clearly defining the problem itself and the main characteristics that the models which are considered online must comply with. We also introduce a novel metric: the Instantaneous Accuracy ( $IA$ ). This new metric exhibits an online nature and solves most of the limitations of the previous metrics. We conduct a thorough experimental evaluation on 3 challenging datasets, where the performance of various baseline methods is compared to that of the state-of-the-art. Our results confirm the problems of the previous evaluation protocols, and suggest that an IA-based protocol is more adequate to the online scenario. The baselines models and a development kit with the novel evaluation protocol will be made publicly available.
AB - The Online Action Detection (OAD) problem needs to be revisited. Unlike traditional offline action detection approaches, where the evaluation metrics are clear and well established, in the OAD setting we find very few works and no consensus on the evaluation protocols to be used. In this work we propose to rethink the OAD scenario, clearly defining the problem itself and the main characteristics that the models which are considered online must comply with. We also introduce a novel metric: the Instantaneous Accuracy ( $IA$ ). This new metric exhibits an online nature and solves most of the limitations of the previous metrics. We conduct a thorough experimental evaluation on 3 challenging datasets, where the performance of various baseline methods is compared to that of the state-of-the-art. Our results confirm the problems of the previous evaluation protocols, and suggest that an IA-based protocol is more adequate to the online scenario. The baselines models and a development kit with the novel evaluation protocol will be made publicly available.
KW - Computer vision
KW - deep learning
KW - evaluation
KW - instantaneous accuracy
KW - online action detection
UR - http://www.scopus.com/inward/record.url?scp=85078147641&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2019.2961789
DO - 10.1109/ACCESS.2019.2961789
M3 - Article
SN - 2169-3536
VL - 8
SP - 5139
EP - 5146
JO - IEEE Access
JF - IEEE Access
M1 - 8939455
ER -