Skip to content

Latest commit

 

History

History
81 lines (67 loc) · 2.33 KB

TE.md

File metadata and controls

81 lines (67 loc) · 2.33 KB
Parameters for testing

The parameters for test_KVNet.py:

exp_name: The name for the experiment. The results will be saved in ../results/${exp_name}

sigma_soft_max: the sigma value used for getting the DPV. Should be the same as during training.

t_win: the time window radius. The time window will be centered around the current frame

d_min, d_max : the minimal and maximal depth values

feature_dim: PSM feature dimension in DNet, should be the same as in the training session

dataset: the dataset name. Should one of scanNet, 7scenes, kitti

dataset_path: the path to the specified dataset

split_file: the spilt txt file, specifying which scenes/videos to use.

model_path: the path to the trained model

Test on ScanNet

Suppose the decoded ScanNet dataset, with 5-frame interval, is at /datasets/scan-net-5-frame, to test on ScanNet dataset:

CUDA_VISIBLE_DEVICES=0 python3 test_KVNet.py \
 		--exp_name te_scannet/ \
 		--sigma_soft_max 10\
 		--t_win 2 \
 		--d_min .1 \
 		--d_max 5 \
 		--feature_dim 64 \
 		--ndepth 64 \
 		--dataset scanNet \
        --dataset_path /datasets/scan-net-5-frame \
        --split_file ./mdataloader/scanNet_split/scannet_val.txt \
 		--model_path ./saved_models/kvnet_scannet.tar

The results will be saved at ../results/te_scannet.

Test on KITTI

Suppose the KITTI dataset is at /datasets/kitti, and the folders are organized as

/datasets/kitti/rawdata
/datasets/kitti/train
/datasets/kitti/val

to test on KITTI dataset:

CUDA_VISIBLE_DEVICES=0 python3 test_KVNet.py \
 		--exp_name te_kitti/ \
 		--sigma_soft_max 10\
 		--t_win 2 \
 		--d_min 1 \
 		--d_max 60 \
 		--feature_dim 64 \
 		--ndepth 64 \
 		--dataset kitti \
        --dataset_path /datasets/kitti \
        --split_file ./mdataloader/kitti_split/test_eigen.txt \
 		--model_path ./saved_models/kvnet_kitti.tar

The results will be saved at ../results/te_kitti.

Test on 7Scenes

Suppose the 7Scenes dataset is at /datasets/7scenes,

CUDA_VISIBLE_DEVICES=0 python3 test_KVNet.py \
 		--exp_name te_7scenes/ \
 		--sigma_soft_max 10\
 		--t_win 2 \
 		--d_min .1 \
 		--d_max 5 \
 		--feature_dim 64 \
 		--ndepth 64 \
 		--dataset 7scenes \
        --dataset_path /datasets/7scenes \
 		--model_path ./saved_models/kvnet_scannet.tar

The results will be saved at ../results/te_7scenes.