基于张量特征的小样本图像快速分类方法
|
|
张艳莎, 冯夫健, 王杰, 潘凤, 谭棉, 张再军, 王林
|
Tensor feature⁃based faster classification network for few⁃shot learning
|
|
Yansha Zhang, Fujian Feng, Jie Wang, Feng Pan, Mian Tan, Zaijun Zhang, Lin Wang
|
|
表2 推理时间和分类准确率的对比
|
Table 2 Comparison of inference time and classification accuracy
|
|
Generate B tensor feature per class | Inference times (s) | Classification accuracy |
|---|
| 1⁃shot | 5⁃shot | 1⁃shot | 5⁃shot |
|---|
| 179 s (214 s) | 218 s (242 s) | 62.65±0.84 (62.43±0.85) | 78.97±0.64 (79.33±0.62) | | 177 s (215 s) | 218 s (244 s) | 63.25±0.83 (63.57±0.87) | 79.00±0.64 (79.94±0.60) | | 181 s (215 s) | 219 s (244 s) | 63.59±0.83 (64.42±0.83) | 79.34±0.64 (79.77±0.61) | | 179 s (216 s) | 219 s (263 s) | 63.66±0.83 (63.90±0.81) | 79.21±0.64 (79.64±0.60) | | 182 s (223 s) | 221 s (249 s) | 63.60±0.83 (63.44±0.86) | 79.08±0.63 (79.43±0.63) | | 186 s (239 s) | 223 s (267 s) | 63.99±0.80 (64.49±0.84) | 79.95±0.61 (79.27±0.60) | | 238 s (426 s) | 268 s (443 s) | 63.59±0.83 (64.25±0.85) | 79.32±0.63 (79.71±0.62) | | 354 s (694 s) | 376 s (715 s) | 63.60±0.83 (63.76±0.87) | 79.10±0.63 (79.52±0.62) |
|
|
|