Fine Grained Image Classification On Oxford
评估指标
Accuracy
FLOPS
PARAMS
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | Accuracy | FLOPS | PARAMS |
|---|---|---|---|
| resnet-strikes-back-an-improved-training | 97.9% | 4.1 | 24M |
| grafit-learning-fine-grained-image | 99.1% | - | - |
| autoformer-searching-transformers-for-visual | - | - | - |
| spinalnet-deep-neural-network-with-gradual-1 | 99.30% | - | - |
| a-comprehensive-study-on-torchvision-pre | 98.29 | - | - |
| pairwise-confusion-for-fine-grained-visual | 93.65% | - | - |
| resmlp-feedforward-networks-for-image | 97.4% | - | - |
| neural-architecture-transfer | 98.1 | 250M | 3.7M |
| an-evolutionary-approach-to-dynamic | 99.61% | - | - |
| transformer-in-transformer | 99.0% | - | 65.6M |
| a-comprehensive-study-on-torchvision-pre | 98.36 | - | - |
| neural-architecture-transfer | - | 152M | 3.3M |
| large-scale-learning-of-general-visual | 99.30% | - | - |
| resmlp-feedforward-networks-for-image | 97.9% | - | - |
| escaping-the-big-data-paradigm-with-compact | - | 15G | 22.5M |
| autoaugment-learning-augmentation-policies | 95.36% | - | - |
| fine-grained-visual-classification-via-2 | 99.64% | - | - |
| neural-architecture-transfer | 97.9 | 195M | 3.4M |
| fixing-the-train-test-resolution-discrepancy | 95.7% | - | - |
| compounding-the-performance-improvements-of | 98.9% | - | - |
| large-scale-learning-of-general-visual | 99.63% | - | - |
| tresnet-high-performance-gpu-dedicated | 99.1% | - | - |
| sr-gnn-spatial-relation-aware-graph-neural | 97.9% | 9.8 | 30.9 |
| neural-architecture-transfer | 98.3 | 400M | 4.2M |
| training-data-efficient-image-transformers | 98.8% | - | 86M |