As human beings utilize computing technologies to mediate multiple aspects of their lives, cyberbullying has grown as an important societal challenge. Cyberbullying may lead to deep psychiatric and emotional disorders for those affected. Hence, there is an urgent need to devise automated methods for cyberbullying detection and prevention. While recent cyberbullying detection efforts have defined sophisticated text processing methods for cyberbullying detection, there are as yet few efforts that leverage visual data processing to automatically detect cyberbullying. Based on early analysis of a public, labeled cyberbullying dataset, we report that visual features complement textual features in cyberbullying detection and can help improve predictive results.