Depth-aware Video Panoptic Segmentation on SemKITTI-DVPS (val)
58.5DVPQPolyphonicFormer
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| PolyphonicFormerlambda=0.50, k=1, Backbone=Swin-B2021.12 | 58.5 | 55.1 | 61 | |
| PolyphonicFormerlambda=0.25, k=1, Backbone=Swin-B2021.12 | 56.3 | 54 | 57.9 | |
| ViP-DeepLablambda=0.50, k=12020.12 | 54.7 | 46.4 | 60.6 | |
| PolyphonicFormerlambda=Average, k=1, Backbone=Swin-B2021.12 | 52.2 | 50.1 | 53.8 | |
| ViP-DeepLablambda=0.25, k=12020.12 | 52 | 44.8 | 57.3 | |
| PolyphonicFormerlambda=0.50, k=5, Backbone=Swin-B2021.12 | 52 | 42.3 | 59.1 | |
| ViP-DeepLablambda=0.50, k=52020.12 | 51.5 | 41 | 59.1 | |
| ViP-DeepLablambda=0.50, k=Average2020.12 | 51.4 | 40.7 | 59.1 | |
| PolyphonicFormerlambda=0.50, k=10, Backbone=Swin-B2021.12 | 50.6 | 39.9 | 58.5 | |
| ViP-DeepLablambda=0.50, k=102020.12 | 50.1 | 38.5 | 58.5 | |
| PolyphonicFormerlambda=0.50, k=20, Backbone=Swin-B2021.12 | 49.9 | 38.6 | 58 | |
| PolyphonicFormerlambda=0.25, k=5, Backbone=Swin-B2021.12 | 49.7 | 41.1 | 56 | |
| ViP-DeepLablambda=0.50, k=202020.12 | 49.2 | 36.9 | 58.2 | |
| ViP-DeepLablambda=Average, k=12020.12 | 48.9 | 42 | 53.9 | |
| ViP-Deeplablambda=Average, k=12021.12 | 48.9 | 42 | 53.9 | |
| ViP-DeepLablambda=0.25, k=52020.12 | 48.8 | 39.4 | 55.7 | |
| ViP-DeepLablambda=0.25, k=Average2020.12 | 48.7 | 39.2 | 55.7 | |
| PolyphonicFormerlambda=0.25, k=10, Backbone=Swin-B2021.12 | 48.4 | 38.7 | 55.5 | |
| PolyphonicFormerlambda=0.25, k=20, Backbone=Swin-B2021.12 | 47.7 | 37.6 | 55 | |
| ViP-DeepLablambda=0.25, k=102020.12 | 47.4 | 37 | 55.1 | |
| ViP-DeepLablambda=0.25, k=202020.12 | 46.6 | 35.6 | 54.7 | |
| PolyphonicFormerlambda=Average, k=Average, Backbone=Swin-B, FLOPs=402G2021.12 | 46.4 | 39 | 51.8 | |
| ViP-DeepLablambda=Average, k=52020.12 | 45.8 | 36.9 | 52.3 | |
| ViP-Deeplablambda=Average, k=52021.12 | 45.8 | 36.9 | 52.3 | |
| ViP-DeepLablambda=Average, k=Average2020.12 | 45.6 | 36.6 | 52.2 | |
| PolyphonicFormerlambda=Average, k=5, Backbone=Swin-B2021.12 | 45.6 | 37.2 | 51.7 | |
| ViP-Deeplablambda=Average, k=Average, FLOPs=2,267G2021.12 | 45.6 | 36.6 | 52.2 | |
| ViP-DeepLablambda=Average, k=102020.12 | 44.4 | 34.6 | 51.6 | |
| ViP-Deeplablambda=Average, k=102021.12 | 44.4 | 34.6 | 51.6 | |
| PolyphonicFormerlambda=Average, k=10, Backbone=Swin-B2021.12 | 44.2 | 34.9 | 51.1 | |
| ViP-DeepLablambda=Average, k=202020.12 | 43.4 | 33 | 51.1 | |
| PolyphonicFormerlambda=Average, k=20, Backbone=Swin-B2021.12 | 43.4 | 33.8 | 50.6 | |
| ViP-Deeplablambda=Average, k=202021.12 | 43.4 | 33 | 51.1 | |
| PolyphonicFormerlambda=0.10, k=1, Backbone=Swin-B2021.12 | 41.8 | 41.1 | 42.4 | |
| ViP-DeepLablambda=0.10, k=12020.12 | 40 | 34.7 | 43.8 | |
| ViP-DeepLablambda=0.10, k=52020.12 | 37.1 | 30.3 | 42 | |
| ViP-DeepLablambda=0.10, k=Average2020.12 | 36.8 | 30 | 41.9 | |
| ViP-DeepLablambda=0.10, k=102020.12 | 35.8 | 28.3 | 41.2 | |
| PolyphonicFormerlambda=0.10, k=5, Backbone=Swin-B2021.12 | 35.1 | 28.2 | 40.1 | |
| ViP-DeepLablambda=0.10, k=202020.12 | 34.5 | 26.5 | 40.4 | |
| PolyphonicFormerlambda=0.10, k=10, Backbone=Swin-B2021.12 | 33.7 | 26 | 39.3 | |
| PolyphonicFormerlambda=0.10, k=20, Backbone=Swin-B2021.12 | 33 | 25.1 | 38.7 |