Flow-Guided Feature Aggregation for Video Object …
aggregation may even deteriorate the performance, as elab-orated in Table 1 (b) later. This suggests that it is critical to model the motion during learning. In this work, we propose flow-guided feature aggrega-tion (FGFA). As illustrated in Figure 1, the feature extrac-tion network is applied on individual frames to produce the per-frame ...
Đọc thêm