Abstract. Traditional video understanding tasks include human action recognition and actor/object semantic segmentation. However, the combined task of providing semantic segmentation for difffferent actor classes simultaneously with their action class remains a challenging but necessary task for many applications. In this work, we propose a new end-to-end architecture for tackling this task in videos. Our model effffectively leverages multiple input modalities, contextual information, and multitask learning in the video to directly output semantic segmentations in a single unifified framework. We train and benchmark our model on the Actor-Action Dataset (A2D) for joint actor-action semantic segmentation, and demonstrate state-of-the-art performance for both segmentation and detection. We also perform experiments verifying our approach improves performance for zero-shot recognition, indicating generalizability of our jointly learned feature space