Object Detection On Pascal Voc 2007
Metriken
MAP
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | MAP |
|---|---|
| objects-as-points | 80.7% |
| spatial-pyramid-pooling-in-deep-convolutional | 60.9% |
| learning-visual-representations-for-transfer-1 | 74.37% |
| yolo9000-better-faster-stronger | 78.6% |
| ultra-efficient-on-device-object-detection-on | 42.3% |
| training-region-based-object-detectors-with | 78.9% |
| thundernet-towards-real-time-generic-object | 78.6% |
| hierarchical-shot-detector | 83.0% |
| deformable-part-models-are-convolutional | 45.2% |
| random-erasing-data-augmentation | 76.2% |
| hierarchical-shot-detector | 81.7% |
| inner-iou-more-effective-intersection-over | - |
| blitznet-a-real-time-deep-network-for-scene | 81.5% |
| rich-feature-hierarchies-for-accurate-object | 58.5% |
| ssd-single-shot-multibox-detector | 81.6% |
| multi-modal-transformers-excel-at-class | 84.16% |
| softer-nms-rethinking-bounding-box-regression | 71.6% |
| fast-r-cnn | 70.0% |
| femtodet-an-object-detection-baseline-for | 22.90% |
| self-knowledge-distillation-a-simple-way-for | 79.7% |
| simple-copy-paste-is-a-strong-data | 89.3% |
| localize-to-classify-and-classify-to-localize | 81.5% |
| dpnet-dual-path-network-for-real-time-object | 79.2% |
| eeea-net-an-early-exit-evolutionary-neural | 81.8% |
| a-fast-rcnn-hard-positive-generation-via | 74.2% |
| couplenet-coupling-global-structure-with | 82.7% |
| subcategory-aware-convolutional-neural | 68.5% |
| you-only-look-once-unified-real-time-object | 63.4% |
| denet-scalable-real-time-object-detection | 77.1% |
| yolo-former-yolo-shakes-hand-with-vit | 86.01% |