So far, several approaches have been studied to bridge the problem of the Semantic Gap, the bottleneck in image and video retrieval. However, no approach is successful enough to increase retrieval performances significantly. One reason is the lack of understanding the user's interest, a major condition towards adapting results to a user. This is partly due to the lack of appropriate interfaces and the missing knowledge of how to interpret user's actions with these interfaces. In this paper, we propose to study the importance of various implicit indicators of relevance. Furthermore, we propose to investigate how this implicit feedback can be combined with static user profiles towards an adaptive video retrieval model.