Image To Image Translation On Synthia To

mIoU (13 classes)

평가 결과

이 벤치마크에서 각 모델의 성능 결과

비교 표

모델 이름	mIoU (13 classes)
daformer-improving-network-architectures-and	67.4
context-aware-mixup-for-domain-adaptive	59.7
hrda-context-aware-high-resolution-domain	72.4
confidence-regularized-self-training	48.7
domain-adaptation-for-structured-output-via	46.5
advent-adversarial-entropy-minimization-for	48
instance-adaptive-self-training-for	57.0
learning-to-adapt-structured-output-space-for	46.7
constructing-self-motivated-pyramid	53.3
a-curriculum-domain-adaptation-approach-to	29.7
mic-masked-image-consistency-for-context	74.0
curriculum-domain-adaptation-for-semantic	29.0
classes-matter-a-fine-grained-adversarial	52.5
category-anchor-guided-unsupervised-domain	44.5
procst-boosting-semantic-segmentation-using	68.2
smoothing-matters-momentum-transformer-for	66.3
all-about-structure-adapting-structural	41.5
pipa-pixel-and-patch-wise-self-supervised	74.8
class-balanced-pixel-level-self-labeling-for	65.3
dada-depth-aware-domain-adaptation-in	49.8
sepico-semantic-guided-pixel-contrast-for	71.4
context-aware-mixup-for-domain-adaptive	69.2
bidirectional-learning-for-domain-adaptation	51.4
sliced-wasserstein-discrepancy-for	48.1
prototypical-pseudo-label-denoising-and	62.0
cross-region-domain-adaptation-for-class	63.7
fcns-in-the-wild-pixel-level-adversarial-and	20.2
learning-to-adapt-structured-output-space-for	45.9