Monocular Depth Estimation On Nyu Depth V2

Métriques

Delta u003c 1.25

Delta u003c 1.25^2

Delta u003c 1.25^3

RMSE

absolute relative error

log 10

Résultats

Résultats de performance de divers modèles sur ce benchmark

Tableau comparatif

Nom du modèle	Delta u003c 1.25	Delta u003c 1.25^2	Delta u003c 1.25^3	RMSE	absolute relative error	log 10
unik3d-universal-camera-monocular-3d	0.989	0.998	1.000	0.173	0.044	0.019
global-local-path-networks-for-monocular	0.915	0.988	0.997	0.344	0.098	0.042
depth-map-decomposition-for-monocular-depth	0.907	0.986	0.997	0.362	0.100	0.043
structure-attentioned-memory-network-for	-	-	-	0.604	-	-
zero-shot-metric-depth-with-a-field-of-view	0.953	0.989	0.996	0.296	0.072	0.031
primedepth-efficient-monocular-depth	0.977	-	-	-	0.046	-
hybriddepth-robust-depth-fusion-for-mobile-ar	0.988	1.000	1.000	0.128	0.026	-
unsupervised-depth-learning-in-challenging	0.820	0.956	-	0.532	0.138	0.059
p3depth-monocular-depth-estimation-with-a	0.898	0.981	0.996	0.356	0.104	0.043
vision-transformers-for-dense-prediction	0.904	0.988	0.994	0.357	0.110	0.045
metric3d-v2-a-versatile-monocular-geometric-1	0.989	0.998	1.000	0.183	0.047	0.020
dinov2-learning-robust-visual-features	0.9497	0.996	0.9994	0.279	0.0907	0.0371
localbins-improving-depth-estimation-by	0.91	0.986	0.997	0.351	0.098	0.042
futuredepth-learning-to-predict-the-future	0.981	0.996	0.999	0.233	0.063	0.027
on-deep-learning-techniques-to-boost	-	-	-	0.429	-	-
a-two-streamed-network-for-estimating-fine	-	-	-	0.635	-	-
metric3d-towards-zero-shot-metric-3d	0.944	0.986	0.995	0.310	0.083	0.035
depth-map-decomposition-for-monocular-depth	0.913	0.987	0.998	0.355	0.098	0.042
nddepth-normal-distance-assisted-monocular	0.936	0.991	0.998	0.311	0.087	0.038
unidepth-universal-monocular-metric-depth	0.984	0.997	0.999	0.201	0.058	0.024
repurposing-diffusion-based-image-generators	0.964	0.991	0.998	0.224	0.055	0.024
unidepthv2-universal-monocular-metric-depth	0.988	0.998	1.000	0.180	0.046	0.020
index-network	-	-	-	0.565	-	-
nvs-monodepth-improving-monocular-depth	-	-	-	0.331	-	-
monocular-depth-estimation-using-diffusion	0.946	0.987	0.996	0.314	0.074	0.032
new-crfs-neural-window-fully-connected-crfs-1	0.922	0.992	0.998	0.334	0.095	0.041
va-depthnet-a-variational-approach-to-single	0.937	0.992	0.999	0.304	0.086	0.037
cutdepth-edge-aware-data-augmentation-in	0.899	0.985	0.997	0.375	0.104	0.044
deep-ordinal-regression-network-for-monocular	-	-	-	0.509	-	-
depthformer-multiscale-vision-transformer-for	0.913	0.988	0.997	0.345	0.100	0.042
analysis-of-nan-divergence-in-training	0.9361	0.9916	0.9981	0.3046	0.0864	0.0365
d-net-a-generalised-and-optimised-deep	0.919	0.988	0.997	0.354	0.095	0.041
harnessing-diffusion-models-for-visual	0.976	0.997	0.999	0.223	0.061	0.027
improving-deep-regression-with-ordinal	0.932	-	-	0.321	0.089	0.039
fast-neural-architecture-search-of-compact	-	-	-	0.526	-	-
mesa-masked-geometric-and-supervised-pre	0.964	0.995	0.999	0.238	0.066	0.029
fine-tuning-image-conditional-diffusion	0.966	-	-	-	0.052	-
zoedepth-zero-shot-transfer-by-combining	0.955	0.995	0.999	0.270	0.075	0.032
revealing-the-dark-secrets-of-masked-image	0.949	0.994	0.999	0.287	0.083	0.035
prompt-guided-transformer-for-multi-task	-	-	-	0.5468	-	-
prompt-guided-transformer-for-multi-task	-	-	-	0.59	-	-
190508598	-	-	-	0.496	-	-
generating-and-exploiting-probabilistic	-	-	-	0.536	-	-
ddp-diffusion-model-for-dense-visual	0.921	0.990	0.998	0.329	0.094	0.040
adabins-depth-estimation-using-adaptive-bins	0.903	0.984	0.997	0.364	0.103	0.044
polymax-general-dense-prediction-with-mask	0.969	0.9958	0.999	0.25	0.067	0.029
pattern-affinitive-propagation-across-depth-1	-	-	-	0.497	-	-
from-big-to-small-multi-scale-local-planar	-	-	0.995	0.392	-	-
unleashing-text-to-image-diffusion-models-for-1	0.964	0.995	0.999	0.254	0.069	0.030
large-scale-monocular-depth-estimation-in-the	0.931	0.986	0.996	0.364	0.080	0.033
ecodepth-effective-conditioning-of-diffusion	0.978	0.997	0.999	0.218	0.059	0.026
grin-zero-shot-metric-depth-with-pixel-level	-	-	-	0.251	0.051	-
predicting-depth-surface-normals-and-semantic	-	-	-	0.641	-	-
irondepth-iterative-refinement-of-single-view	0.910	0.985	0.997	0.352	0.101	0.043
attention-based-context-aggregation-network	-	-	-	0.496	-	-
real-time-joint-semantic-segmentation-and	-	-	-	0.565	-	-
sdc-depth-semantic-divide-and-conquer-network	-	-	-	0.497	-	-
iebins-iterative-elastic-bins-for-monocular-1	0.936	0.992	0.998	0.314	0.087	0.038
binsformer-revisiting-adaptive-bins-for	0.925	0.989	0.997	0.330	0.094	0.040
revisiting-single-image-depth-estimation	-	-	-	0.530	-	-
all-in-tokens-unifying-output-space-of-visual	0.954	0.994	0.999	0.275	0.076	0.033
structure-aware-residual-pyramid-network-for	-	-	-	0.514	-	-
fast-neural-architecture-search-of-compact	-	-	-	0.523	-	-
enforcing-geometric-constraints-of-virtual	0.875	0.976	0.989	0.416	0.111	0.048
monocular-depth-estimation-using-laplacian	0.895	0.983	0.996	0.384	0.105	0.045
high-quality-monocular-depth-estimation-via	-	-	-	0.465	-	-
primedepth-efficient-monocular-depth	0.966	-	-	-	0.058	-
idisc-internal-discretization-for-monocular	-	0.993	0.999	-	0.086	-
text-image-alignment-for-diffusion-based	0.976	0.997	0.999	0.225	0.062	0.027
attention-attention-everywhere-monocular	0.929	0.991	0.998	0.322	0.090	0.039
depth-anything-unleashing-the-power-of-large	0.984	0.998	1.000	0.206	0.056	0.024
depthmaster-taming-diffusion-models-for	0.972	-	-	-	0.050	-
single-image-depth-estimation-trained-via-1	-	-	-	0.575	-	-
fast-neural-architecture-search-of-compact	-	-	-	0.525	-	-
neural-video-depth-stabilizer	0.9493	0.991	0.997	0.282	0.072	0.031
depthformer-exploiting-long-range-correlation	0.921	0.989	0.998	0.339	0.096	0.041
urcdc-depth-uncertainty-rectified-cross	0.933	0.992	0.998	0.316	0.088	0.038
monocular-depth-estimation-using-relative	-	-	-	0.538	-	-
multi-scale-continuous-crfs-as-sequential	-	-	-	0.586	-	-
learning-to-recover-3d-scene-shape-from-a	0.916	-	-	-	0.09	-
inverted-pyramid-multi-task-transformer-for	-	-	-	0.5183	-	-
focal-wnet-an-architecture-unifying	0.875	0.980	0.995	0.398	0.116	0.048
scaledepth-decomposing-metric-depth	0.957	0.994	0.999	0.267	0.074	0.032
evp-enhanced-visual-perception-using-inverse	0.976	0.997	0.999	0.224	0.061	0.027
distill-any-depth-distillation-creates-a	0.981	-	-	-	0.043	-