Abstract We address semantic video object segmentation via a novel cross-granularity hierarchical graphical model to integrate tracklet and object proposal reasoning with superpixel labeling. Tracklet characterizes varying spatial-temporal relations of video object which, however, quite often suffers from sporadic local outliers. In order to acquire highquality tracklets, we propose a transductive inference model which is capable of calibrating shortrange noisy object tracklets with respect to longrange dependencies and high-level context cues. In the center of this work lies a new paradigm of semantic video object segmentation beyond modeling appearance and motion of objects locally, where the semantic label is inferred by jointly exploiting multi-scale contextual information and spatialtemporal relations of video object. We evaluate our method on two popular semantic video object segmentation benchmarks and demonstrate that it advances the state-of-the-art by achieving superior accuracy performance than other leading methods