Action Segmentation On 50 Salads 1

Métriques

Acc

Edit

F1@10%

F1@25%

F1@50%

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Acc	Edit	F1@10%	F1@25%	F1@50%	Paper Title	Repository
BaFormer	89.5	84.2	89.3	88.4	83.9	Efficient Temporal Action Segmentation via Boundary-aware Query Voting
ASFormer+ASRF	85.9	81.9	85.1	85.4	79.3	ASFormer: Transformer for Action Segmentation
SF-TMN(ASFormer)	89.8	84.4	89.1	88.0	82.9	SF-TMN: SlowFast Temporal Modeling Network for Surgical Phase Recognition	-
ASRF	84.5	79.3	84.9	83.5	77.3	Alleviating Over-segmentation Errors by Detecting Action Boundaries
SSTDA	83.2	75.8	83.0	81.5	73.8	Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation
EUT	87.4	82.9	89.2	87.5	81	Do we really need temporal convolutions in action segmentation?
LTContext	87.7	83.2	89.4	87.7	82.0	How Much Temporal Long-Term Context is Needed for Action Segmentation?
CETNet	86.9	81.7	87.6	86.5	80.1	Cross-Enhancement Transformer for Action Segmentation
MS-TCN	80.7	67.9	76.3	74.0	64.5	MS-TCN: Multi-Stage Temporal Convolutional Network for Action Segmentation
DA	83.2	75.2	82.0	80.1	72.5	Action Segmentation with Mixed Temporal Domain Adaptation	-
MS-TCN++(sh)	82.2	70.7	78.7	76.6	68.3	MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation
DS-TCN	80.0	70.0	77.0	74.43	65.78	Depthwise Separable Temporal Convolutional Network for Action Segmentation	-
DTGRM	80	72	79.1	75.9	66.1	Temporal Relational Modeling with Self-Supervision for Action Segmentation
UVAST	87.4	83.9	89.1	87.6	81.7	Unified Fully and Timestamp Supervised Temporal Action Segmentation via Sequence to Sequence Translation
G2L (MS-TCN)	82.2	73.4	80.3	78	69.8	Global2Local: Efficient Structure Search for Video Action Segmentation
ETSN	82.0	78.8	85.2	83.9	75.4	Efficient Two-Step Networks for Temporal Action Segmentation
C2F-TCN	84.9	76.4	84.3	81.8	72.6	Coarse to Fine Multi-Resolution Temporal Convolutional Network
BCN	84.4	74.3	82.3	81.3	74	Boundary-Aware Cascade Networks for Temporal Action Segmentation
Br-Prompt+ASPnet (RGB, flow, accelerometer)	91.4	87.5	92.7	91.6	88.5	ASPnet: Action Segmentation With Shared-Private Representation of Multiple Data Sources	-
AUL	77.9	77.0	84.4	81.3	67.1	Is Weakly-supervised Action Segmentation Ready For Human-Robot Interaction? No, Let's Improve It With Action-union Learning

0 of 28 row(s) selected.