Attend What You Need: Motion-Appearance Synergistic Networks for Video Question Answering

Open in new window