mAP IOU@0.5: 63.5mAP IOU@0.3: 75.5mAP IOU@0.4: 70.8mAP IOU@0.6: 50.9
Papers with Code · CC BY-SA 4.0
Zero-Shot Learning
Zero-shot learning (ZSL) is a model's ability to detect classes never seen during training. The condition is that the classes are not known during supervised learning.
Earlier work in zero-shot learning use attributes in a two-step approach to infer unknown classes. In the computer vision context, more recent advances learn mappings from image feature space to semantic space. Other approaches learn non-linear multimodal embeddings. In the modern NLP context, language models can be evaluated on downstream tasks without fine tuning.
Benchmark datasets for zero-shot learning include aPY, AwA, and CUB, among others.
( Image credit: Prototypical Networks for Few shot Learning in PyTorch )
Further readings:
논문
mAP IOU@0.5: 62.53mAP IOU@0.3: 73.29mAP IOU@0.4: 69.73mAP IOU@0.6: 53.36
mAP IOU@0.5: 71.7mAP IOU@0.3: 83.1mAP IOU@0.4: 79.0mAP IOU@0.6: 59.7
mAP IOU@0.5: 71.0mAP IOU@0.3: 82.1mAP IOU@0.4: 77.8mAP IOU@0.6: 59.4
mAP IOU@0.5: 50.2mAP IOU@0.1: 72.1mAP IOU@0.2: 69.8mAP IOU@0.3: 65
mAP IOU@0.5: 56.0mAP IOU@0.3: 69.4mAP IOU@0.4: 64.3mAP IOU@0.6: 46.4
mAP IOU@0.5: 36.9mAP IOU@0.3: 53.5mAP IOU@0.4: 45mAP IOU@0.6: 28.4
mAP IOU@0.5: 31mAP IOU@0.1: 60.1mAP IOU@0.2: 56.7mAP IOU@0.3: 50.1
mAP IOU@0.5: 54.1mAP IOU@0.3: 68.2mAP IOU@0.4: 62.7mAP IOU@0.6: 43.9
mAP IOU@0.5: 44.2mAP IOU@0.3: 60.2mAP IOU@0.4: 54.1mAP IOU@0.6: 32.3
mAP IOU@0.5: 70.4mAP IOU@0.3: 82.9mAP IOU@0.4: 78.0mAP IOU@0.6: 58.5
mAP IOU@0.5: 17.1mAP IOU@0.1: 48.9mAP IOU@0.2: 44.0mAP IOU@0.3: 36.0
mAP IOU@0.5: 60.1mAP IOU@0.3: 74.8mAP IOU@0.4: 69.1mAP IOU@0.6: 46.6
mAP IOU@0.5: 51.9mAP IOU@0.1: 72.5mAP IOU@0.2: 70.9mAP IOU@0.3: 66.5
mAP IOU@0.5: 49.1mAP IOU@0.1: 69.5mAP IOU@0.2: 67.8mAP IOU@0.3: 63.6
mAP IOU@0.5: 57.1mAP IOU@0.3: 70.1mAP IOU@0.4: 64.9mAP IOU@0.6: 45.4
mAP IOU@0.5: 56.9mAP IOU@0.3: 68.9mAP IOU@0.4: 64.0mAP IOU@0.6: 46.3
mAP IOU@0.5: 57.0mAP IOU@0.3: 68.6mAP IOU@0.4: 63.8mAP IOU@0.6: 46.3
mAP IOU@0.5: 28.9mAP IOU@0.1: 54.5mAP IOU@0.2: 51.5mAP IOU@0.3: 44.8
mAP IOU@0.5: 57.1mAP IOU@0.3: 69.2mAP IOU@0.4: 65.0mAP IOU@0.6: 47.8
mAP IOU@0.5: 42.8mAP IOU@0.1: 59.8mAP IOU@0.2: 57.1mAP IOU@0.3: 53.2
mAP IOU@0.5: 53.8mAP IOU@0.3: 62.8mAP IOU@0.4: 59.5mAP IOU@0.6: 43.6
mAP IOU@0.5: 19mAP IOU@0.1: 47.7mAP IOU@0.2: 43.5mAP IOU@0.3: 36.3
mAP IOU@0.5: 71.8mAP IOU@0.3: 82.8mAP IOU@0.4: 78.9mAP IOU@0.6: 60.5
mAP IOU@0.5: 53.5mAP IOU@0.1: 74.02mAP IOU@0.2: 72.29mAP IOU@0.3: 69.1
mAP IOU@0.5: 25.6mAP IOU@0.1: 54mAP IOU@0.2: 50.9mAP IOU@0.3: 44.1
mAP IOU@0.5: 52.4mAP IOU@0.3: 66.7mAP IOU@0.4: 60.4mAP IOU@0.6: 41.0
mAP IOU@0.5: 29.6mAP IOU@0.1: 62.3mAP IOU@0.3: 46.8mAP IOU@0.7: 9.7