Abstract:Summary TRECVID Multimedia Event Detection offers an interesting but very challenging task in detecting high- level complex events (Figure 1) in user-generated videos. In this paper, we will present an overview and comparative analysis of our results, which achieved top performance among all 45 submissions in TRECVID 2010. Our aim is to answer the following questions. What kind of feature is more effective for multimedia event detection? Are features from different feature modalities (e.g., audio and visual) complementary for event detection? Can we benefit from generic concept detection of background scenes, human actions, and audio concepts? Are sequence matching and event-specific object detectors critical? Our findings indicate that spatial-temporal feature is very effective for event detection, and it's also very complementary to other features such as static SIFT and audio features. As a result, our baseline run combining these three features already achieves very impressive results, with a mean minimal normalized cost (MNC) of 0.586. Incorporating the generic concept detectors using a graph diffusion algorithm provides marginal gains (mean MNC 0.579). Sequence matching with Earth Mover's Distance (EMD) further improves the results (mean MNC 0.565). The event-specific detector (batter), however, didn't prove useful from our current re-ranking tests. We conclude that it is important to combine strong complementary features from multiple modalities for multimedia event detection, and cross-frame matching is helpful in coping with temporal order variation. Leveraging contextual concept detectors and foreground activities remains a very attractive direction requiring further research.

Event Detection and Tracking Based on Improved Incremental K-Means and Transductive SVM

Object Tracking Method Based on Incrementally Updating Linear Discriminant Subspace

A System for Detecting and Tracking Internet News Event

A NEP-SVM based simulation system for event tracking

A System for Event Detection and Tracking Based on Constructive-Competition Clustering and KNNFL

Moving Object Detection Based on SIFT Features Matching and K-means Clustering

Spatiotemporal latent semantic cues for moving people tracking

The improved spatial histogram features and integrated system for vehicle detection

Study on Network Intrusion Detection Based on TSVM

BIT @ TRECVid 2013 : Surveillance Event Detection

Informedia@TRECVID 2011: Surveillance Event Detection.

Adaptive Online Event Detection in News Streams.

Exciting Event Detection in Broadcast Soccer Video with Mid-Level Description and Incremental Learning

Topic Detection and Tracking Based on Windowed DBSCAN and Parallel KNN

Crowd Event Detection Based on Motion Vector Intersection Points

Effective video event detection via subspace projection

Feature Tracking Utilizing a Maximum-Weight Independent Set Problem

Clustering Based Topic Events Detection On Text Stream

An Improved Discriminative Tracking Algorithm Based on Spatial Information and Model Updating

Columbia-UCF TRECVID2010 Multimedia Event Detection: Combining Multiple Modalities, Contextual Concepts, and Temporal Matching.

Video Event Detection: From Subvolume Localization to Spatiotemporal Path Search