[ot][ml] Current Video Object Segmentation Models

[Undescribed Horrific Abuse, One Victim & Survivor of Many]

[there are mistakes in last post. attempting to halt posts to this
thread to curb future mistakespam. mistake update: 1. it is the full
leaderboard, it just uses javascript. 2. hqtrack would be second place
because #1 and #2 are from the same group. 3. the winning model is
DMAOT [possible mistake: from the same people who did SAM-Track?]]