HyperAI
HyperAI超神经
首页
算力平台
文档
资讯
论文
教程
数据集
百科
SOTA
LLM 模型天梯
GPU 天梯
顶会
开源项目
全站搜索
关于
中文
HyperAI
HyperAI超神经
Toggle sidebar
全站搜索…
⌘
K
Command Palette
Search for a command to run...
首页
SOTA
视觉物体跟踪
Visual Object Tracking On Trackingnet
Visual Object Tracking On Trackingnet
评估指标
Accuracy
Normalized Precision
Precision
评测结果
各个模型在此基准测试上的表现结果
Columns
模型名称
Accuracy
Normalized Precision
Precision
Paper Title
Repository
MCITrack-L384
87.9
92.1
89.2
Exploring Enhanced Contextual Information for Video-Level Object Tracking
MCITrack-B224
86.3
90.9
86.1
Exploring Enhanced Contextual Information for Video-Level Object Tracking
ODTrack-L
86.1
-
-
ODTrack: Online Dense Temporal Token Learning for Visual Tracking
MixViT-L(ConvMAE)
86.1
90.3
86.0
MixFormer: End-to-End Tracking with Iterative Mixed Attention
ARTrackV2-L
86.1
90.4
86.2
ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe
LoRAT-g-378
86.0
90.2
86.1
Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance
ARTrack-L
85.6
89.6
86.0
Autoregressive Visual Tracking
-
LoRAT-L-378
85.6
89.7
85.4
Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance
SeqTrack-L384
85.5
89.8
85.8
Unified Sequence-to-Sequence Learning for Single- and Multi-Modal Visual Object Tracking
UNINEXT-H
85.4
89.0
86.4
Universal Instance Perception as Object Discovery and Retrieval
SAMURAI-L
85.3
-
-
SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory
ODTrack-B
85.1
-
-
ODTrack: Online Dense Temporal Token Learning for Visual Tracking
TATrack-L
85.0
89.3
84.5
Target-Aware Tracking with Long-term Context Attention
HIPTrack
84.5
89.1
83.8
HIPTrack: Visual Tracking with Historical Prompts
SwinTrack-B-384
84
88.2
83.2
SwinTrack: A Simple and Strong Baseline for Transformer Tracking
OSTrack-384
83.9
88.5
83.2
Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework
MixFormer-L
83.9
88.9
83.1
MixFormer: End-to-End Tracking with Iterative Mixed Attention
NeighborTrack-OSTrack
83.79
88.30
-
NeighborTrack: Improving Single Object Tracking by Bipartite Matching with Neighbor Tracklets
MixFormerV2-B
83.4
88.1
81.6
-
-
MITS
83.4
88.9
84.6
Integrating Boxes and Masks: A Multi-Object Framework for Unified Visual Tracking and Segmentation
0 of 37 row(s) selected.
Previous
Next