Multispectral Object Detection On Flir 1
Metriken
mAP
mAP50
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
Modellname | mAP | mAP50 |
---|---|---|
multimodal-object-detection-by-channel | 37.9% | 75.5% |
cross-modality-fusion-transformer-for | - | 77.7% |
when-pedestrian-detection-meets-multi-modal | - | 86.4% |
Modell 4 | 50.5% | 85.5% |
unirgb-ir-a-unified-framework-for-visible | 44.1% | 81.4% |
multimodal-object-detection-by-channel | 41.3% | 79.2% |
cross-modality-fusion-transformer-for | - | 73.9% |
cross-modality-fusion-transformer-for | - | 67.8% |
guided-attentive-feature-fusion-for | - | 72.9% |
cmx-cross-modal-fusion-for-rgb-x-semantic | - | 82.2% |
mipa-mixed-patch-infrared-visible-modality | 44.8% | 81.3% |
rgb-x-object-detection-via-scene-specific | 47.1% | 86.16% |
guided-attentive-feature-fusion-for | - | 72.7% |
multimodal-object-detection-by-channel | 37.4% | 74.6% |
removal-and-selection-improving-rgb-infrared | 43.8% | 83.9% |
multispectral-fusion-for-object-detection | - | 71.2% |
multimodal-object-detection-by-channel | 35.8% | - |
multispectral-fusion-for-object-detection | - | 72.4% |