HyperAIHyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAIHyperAI超神経

Visual Object Tracking On Got 10K

評価指標

Average Overlap
Success Rate 0.5
Success Rate 0.75

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Average Overlap
Success Rate 0.5
Success Rate 0.75
Paper TitleRepository
SLT-TransT67.576.860.3Towards Sequence-Level Training for Visual Tracking-
OSTrack-38473.783.270.8Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework-
MCITrack-B22477.988.276.8Exploring Enhanced Contextual Information for Video-Level Object Tracking-
MCITrack-L38480.088.580.2Exploring Enhanced Contextual Information for Video-Level Object Tracking-
DiMP61.171.7-Learning Discriminative Model Prediction for Tracking-
SeqTrack-L38474.881.972.2Unified Sequence-to-Sequence Learning for Single- and Multi-Modal Visual Object Tracking-
Siam R-CNN64.972.8-Siam R-CNN: Visual Tracking by Re-Detection-
MITS80.489.875.8Integrating Boxes and Masks: A Multi-Object Framework for Unified Visual Tracking and Segmentation-
STMTrack64.273.757.5STMTrack: Template-free Visual Tracking with Space-time Memory Networks-
ATOM61.074.2-ATOM: Accurate Tracking by Overlap Maximization-
ARTrackV2-L79.587.879.6ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe-
TRASFUST61.772.9-Tracking-by-Trackers with a Distilled and Reinforced Model-
NeighborTrack-OSTrack75.785.7273.3NeighborTrack: Improving Single Object Tracking by Bipartite Matching with Neighbor Tracklets-
ARTrack-L78.587.477.8Autoregressive Visual Tracking
FEAR-M62.3--FEAR: Fast, Efficient, Accurate and Robust Visual Tracker-
SAMURAI-L81.792.276.9SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory-
TATrack-L-GOT76.685.773.4Target-Aware Tracking with Long-term Context Attention-
Ocean61.172.1-Ocean: Object-aware Anchor-free Tracking-
MixFormer-1k71.279.965.8MixFormer: End-to-End Tracking with Iterative Mixed Attention-
FEAR-L64.5--FEAR: Fast, Efficient, Accurate and Robust Visual Tracker-
0 of 39 row(s) selected.
Visual Object Tracking On Got 10K | SOTA | HyperAI超神経