Action Segmentation On Gtea 1

评估指标

Acc

Edit

F1@10%

F1@25%

F1@50%

评测结果

各个模型在此基准测试上的表现结果

						Paper Title	Repository
Semantic2Graph	89.8	92.0	95.7	94.2	91.3	Semantic2Graph: Graph-based Multi-modal Feature Fusion for Action Segmentation in Videos	-
FACT	84.5	93.5	96.1	95.6	87.5	FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation	-
SF-TMN(ASFormer)	83.0	88.9	91.9	90.7	83.1	SF-TMN: SlowFast Temporal Modeling Network for Surgical Phase Recognition	-
BaFormer	83.0	88.7	92.0	91.3	83.5	Efficient Temporal Action Segmentation via Boundary-aware Query Voting
DiffAct	82.2	89.6	92.5	91.5	84.7	Diffusion Action Segmentation
DPRN	82.0	90.9	92.9	92.0	82.9	Maximization and restoration: Action segmentation through dilation passing and temporal reconstruction	-
BIT	82.0	92.6	94.8	92.8	82.6	BIT: Bi-Level Temporal Modeling for Efficient Supervised Action Segmentation	-
Br-Prompt+ASFormer	81.2	91.6	94.1	92.0	83.0	Bridge-Prompt: Towards Ordinal Action Understanding in Instructional Videos
C2F-TCN	80.8	86.4	90.3	88.8	77.7	Coarse to Fine Multi-Resolution Temporal Convolutional Network
CETNet	80.3	87.9	91.8	91.2	81.3	Cross-Enhancement Transformer for Action Segmentation
UVAST	80.2	92.1	92.7	91.3	81	Unified Fully and Timestamp Supervised Temporal Action Segmentation via Sequence to Sequence Translation
MS-TCN++	80.1	83.5	88.8	85.7	76.0	MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation
DA	80.0	85.8	90.5	88.4	76.2	Action Segmentation with Mixed Temporal Domain Adaptation	-
BCN	79.8	84.4	88.5	87.1	77.3	Boundary-Aware Cascade Networks for Temporal Action Segmentation	-
SSTDA	79.8	86.2	90.0	89.1	78.0	Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation
ASFormer	79.7	84.6	90.1	88.8	79.2	ASFormer: Transformer for Action Segmentation
MS-TCN++(sh)	79.7	83.0	88.2	86.2	75.9	MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation
MS-TCN	79.2	81.4	87.5	85.4	74.6	MS-TCN: Multi-Stage Temporal Convolutional Network for Action Segmentation
SSTDA + HASR	78.7	87.5	90.9	88.6	76.4	Refining Action Segmentation With Hierarchical Video Representations	-
ETSN	78.2	86.2	91.1	90.0	77.9	Efficient Two-Step Networks for Temporal Action Segmentation	-

0 of 28 row(s) selected.