Continual Generalized Zero-Shot Learning on AWA1 benchmark leaderboard · SOTA2 Research

Continual Generalized Zero-Shot Learning on AWA1

85.01mSA (Seen)CV +mem

Evaluation Results

Method
CV +mem2022.03	85.01	37.49	51.6
CA +mem2022.03	81.86	61.39	69.92
unified feature-generative framework2022.03	79.51	69.13	73.49
CV +mem2022.03	78.56	23.65	35.51
Seq-CADA2022.03	78.12	25.93	47.06
A-CZSL2022.03	78.03	35.38	52.51
DVGR-CZSL2022.03	76.9	42.04	53.38
NM-ZSL2022.03	75.59	60.87	67.44
A-CZSL2022.03	71	24.26	35.75
NM-ZSL2022.03	70.9	37.46	48.75
Seq-CVAE2022.03	70.24	28.36	39.32
unified feature-generative framework2022.03	66.3	50.63	55.59
DVGR-CZSL2022.03	65.1	28.5	38
Tf-GCZSL2022.03	64	38.34	46.16
CA +mem2022.03	62.64	38.41	45.38
Tf-GCZSL2022.03	61.79	57.77	59.72
Seq-CVAE2022.03	59.27	18.24	27.14
Seq-CADA2022.03	51.57	18.02	27.59
EWC2022.03	50.41	—	—
EWC2022.03	42.6	—	—
Seq-fCLSWGAN2022.03	23.39	12.57	16.35
Seq-fCLSWGAN2022.03	20.32	9.25	12.71