From 786f1897defeeea809943b7ad61fc39a43de30ff Mon Sep 17 00:00:00 2001 From: Qunjie Zhou Date: Thu, 11 Jan 2024 22:37:48 +0100 Subject: [PATCH] Update evaluation.md , add relapose command --- docs/evaluation.md | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/docs/evaluation.md b/docs/evaluation.md index 0d02b59..a7dbd6a 100644 --- a/docs/evaluation.md +++ b/docs/evaluation.md @@ -11,6 +11,8 @@ All datasets should be placed under **data/datasets**. Here, we provide brief d - **RobotCar Seasons**: Download the [RobotCar Seasons](https://data.ciirc.cvut.cz/public/projects/2020VisualLocalization/RobotCar-Seasons/) dataset contents and place them under a folder named **RobotCar/**. + - **MegaDepth** and **ScanNet** for relative pose estimation: we followed [LoFTR](https://github.com/zju3dv/LoFTR?tab=readme-ov-file) to setup both datasets for testing, please refer to their repo for details. + ### Prepare Image Pairs To evaluate on visual localization benchmarks, one needs to prepare image pairs that are required by HLoc pipeline in advance. For convenience, we cached the pairs that are extracted by [HLoc](https://github.com/cvg/Hierarchical-Localization) author Paul-Edouard Sarlin. @@ -76,6 +78,17 @@ python -m immatch.eval_hpatches --gpu 0 \ --task 'both' --save_npy \ --root_dir . ``` +### Relative Pose Estimation +To reproduce [AspanFormer results](https://github.com/apple/ml-aspanformer/tree/main?tab=readme-ov-file#evaluation): +``` +# MegaDepth + +python -m immatch.eval_relapose --config 'aspanformer' --benchmark 'megadepth' + +# ScanNet +python -m immatch.eval_relapose --config 'aspanformer' --benchmark 'scannet' +``` + ### Long-term Visual Localization We adopt the public implementation [Hierarchical Localization](https://github.com/cvg/Hierarchical-Localization) to evaluate matches on several long-term visual localization benchmarks, including: