CADA-VAE

Reported on 98 benchmarks across 12 tasks · 1 paper · 54 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology42 results

Zero-Shot LearningonNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Zero-Shot LearningonNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Zero-Shot LearningonNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Zero-Shot LearningonPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Zero-Shot LearningonNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Zero-Shot LearningonNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Zero-Shot LearningonNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (1-shot)· 2018-12-05
69.6
best: 69.9 (MVCN)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (10-shots)· 2018-12-05
80.2
best: 82.2 (MVCN)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (2-shots)· 2018-12-05
73.7
best: 76.4 (MVCN)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (20-shots)· 2018-12-05
80.9
best: 83.3 (DRAGON)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Generalized Few-Shot LearningonAwA2
Per-Class Accuracy (5-shots)· 2018-12-05
78.1
best: 81.2 (MVCN)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Generalized Few-Shot ClassificationonCUB-LT
Long-Tailed Accuracy
57.4
best: 67.7 (DRAGON)
Generalized Few-Shot ClassificationonCUB-LT
Per-Class Accuracy
48.3
best: 60.1 (DRAGON + Bal'Loss)
Generalized Few-Shot ClassificationonAWA-LT
Long-Tailed Accuracy
89.5
best: 94.1 (DRAGON)
Generalized Few-Shot ClassificationonAWA-LT
Per-Class Accuracy
73.5
best: 76.2 (DRAGON + Bal'Loss)
Generalized Few-Shot ClassificationonSUN-LT
Long-Tailed Accuracy
35.1
best: 40.4 (DRAGON)
Generalized Few-Shot ClassificationonSUN-LT
Per-Class Accuracy
32.8
best: 36.1 (DRAGON + Bal'Loss)
Generalized Few-Shot ClassificationonImageNet-LT-d
Per-Class Accuracy
49.3
best: 53.5 (DRAGON + Bal'Loss)
Long-tail LearningonCUB-LT
Long-Tailed Accuracy
57.4
best: 67.7 (DRAGON)
Long-tail LearningonCUB-LT
Per-Class Accuracy
48.3
best: 60.1 (DRAGON + Bal'Loss)
Long-tail LearningonAWA-LT
Long-Tailed Accuracy
89.5
best: 94.1 (DRAGON)
Long-tail LearningonAWA-LT
Per-Class Accuracy
73.5
best: 76.2 (DRAGON + Bal'Loss)
Long-tail LearningonSUN-LT
Long-Tailed Accuracy
35.1
best: 40.4 (DRAGON)
Long-tail LearningonSUN-LT
Per-Class Accuracy
32.8
best: 36.1 (DRAGON + Bal'Loss)
Long-tail LearningonImageNet-LT-d
Per-Class Accuracy
49.3
best: 53.5 (DRAGON + Bal'Loss)
Generalized Few-Shot LearningonCUB
Per-Class Accuracy (2-shots)
59.2
best: 61.6 (MVCN)
Generalized Few-Shot LearningonCUB
Per-Class Accuracy (1-shot)
55.2
best: 57.3 (MVCN)
Generalized Few-Shot LearningonCUB
Per-Class Accuracy (10-shots)
64.9
best: 67.8 (MVCN)
Generalized Few-Shot LearningonCUB
Per-Class Accuracy (20-shots)
66
Generalized Few-Shot LearningonCUB
Per-Class Accuracy (5-shots)
63
best: 65.4 (MVCN)
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (1-shot)
37.8
best: 41 (DRAGON)
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (10-shots)
45.8
best: 48.2 (DRAGON)
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (2-shots)
41.4
best: 43.8 (DRAGON)
Generalized Few-Shot LearningonSUN
Per-Class Accuracy (5-shots)
44.2
best: 46.7 (DRAGON)
Generalized Few-Shot LearningonCUB-LT
Long-Tailed Accuracy
57.4
best: 67.7 (DRAGON)
Generalized Few-Shot LearningonCUB-LT
Per-Class Accuracy
48.3
best: 60.1 (DRAGON + Bal'Loss)
Generalized Few-Shot LearningonAWA-LT
Long-Tailed Accuracy
89.5
best: 94.1 (DRAGON)
Generalized Few-Shot LearningonAWA-LT
Per-Class Accuracy
73.5
best: 76.2 (DRAGON + Bal'Loss)
Generalized Few-Shot LearningonSUN-LT
Long-Tailed Accuracy
35.1
best: 40.4 (DRAGON)
Generalized Few-Shot LearningonSUN-LT
Per-Class Accuracy
32.8
best: 36.1 (DRAGON + Bal'Loss)
Generalized Few-Shot LearningonImageNet-LT-d
Per-Class Accuracy
49.3
best: 53.5 (DRAGON + Bal'Loss)

Computer Vision35 results

VideoonNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
VideoonNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
VideoonNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
VideoonPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
VideoonNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
VideoonNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
VideoonNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Temporal Action LocalizationonNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action LocalizationonNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Image ClassificationonCUB-LT
Long-Tailed Accuracy
57.4
best: 67.7 (DRAGON)
Image ClassificationonCUB-LT
Per-Class Accuracy
48.3
best: 60.1 (DRAGON + Bal'Loss)
Image ClassificationonAWA-LT
Long-Tailed Accuracy
89.5
best: 94.1 (DRAGON)
Image ClassificationonAWA-LT
Per-Class Accuracy
73.5
best: 76.2 (DRAGON + Bal'Loss)
Image ClassificationonSUN-LT
Long-Tailed Accuracy
35.1
best: 40.4 (DRAGON)
Image ClassificationonSUN-LT
Per-Class Accuracy
32.8
best: 36.1 (DRAGON + Bal'Loss)
Image ClassificationonImageNet-LT-d
Per-Class Accuracy
49.3
best: 53.5 (DRAGON + Bal'Loss)
Few-Shot Image ClassificationonCUB-LT
Long-Tailed Accuracy
57.4
best: 67.7 (DRAGON)
Few-Shot Image ClassificationonCUB-LT
Per-Class Accuracy
48.3
best: 60.1 (DRAGON + Bal'Loss)
Few-Shot Image ClassificationonAWA-LT
Long-Tailed Accuracy
89.5
best: 94.1 (DRAGON)
Few-Shot Image ClassificationonAWA-LT
Per-Class Accuracy
73.5
best: 76.2 (DRAGON + Bal'Loss)
Few-Shot Image ClassificationonSUN-LT
Long-Tailed Accuracy
35.1
best: 40.4 (DRAGON)
Few-Shot Image ClassificationonSUN-LT
Per-Class Accuracy
32.8
best: 36.1 (DRAGON + Bal'Loss)
Few-Shot Image ClassificationonImageNet-LT-d
Per-Class Accuracy
49.3
best: 53.5 (DRAGON + Bal'Loss)

Robots7 results

Activity RecognitiononNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Activity RecognitiononNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Activity RecognitiononNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Activity RecognitiononPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Activity RecognitiononNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Activity RecognitiononNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Activity RecognitiononNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784

Natural Language Processing7 results

3D Action RecognitiononNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
3D Action RecognitiononNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
3D Action RecognitiononNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
3D Action RecognitiononPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
3D Action RecognitiononNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
3D Action RecognitiononNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
3D Action RecognitiononNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784

Time Series7 results

Action RecognitiononNTU RGB+D 120
Accuracy (10 unseen classes)· 2018-12-05
59.53
best: 79.99 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action RecognitiononNTU RGB+D 120
Accuracy (24 unseen classes)· 2018-12-05
35.77
best: 65.06 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action RecognitiononNTU RGB+D 120
Random Split Accuracy· 2018-12-05
45.14
best: 71.38 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action RecognitiononPKU-MMD
Random Split Accuracy· 2018-12-05
60.74
best: 87.8 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action RecognitiononNTU RGB+D
Accuracy (12 unseen classes)· 2018-12-05
28.96
best: 56.03 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action RecognitiononNTU RGB+D
Accuracy (5 unseen classes)· 2018-12-05
76.84
best: 86.49 (TDSM)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784
Action RecognitiononNTU RGB+D
Random Split Accuracy· 2018-12-05
60.74
best: 93.17 (PGFA)
SOTA
Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders arXiv:1812.01784