search

yzdad / D-GLSNet

The code of Geo-localization with Transformer-based 2D-3D match Network
37 stars 6 forks source link

satellite image of kitti-360 #3

Open realxuyang opened 11 months ago

realxuyang commented 11 months ago

hello author, how can i get the satellite image of kitti-360? i haven't seen them on their website

yzdad commented 11 months ago

We obtain satellite images through the website provided by Google Map. You can obtain the corresponding map through the following website.

https://gac-geo.googlecnapps.cn/maps/vt?lyrs=s&x={x}&y={y}&z={z} x: Information number of tile X-axis y: Information number of tile Y-axis z: Scale level of tiles

hty1233 commented 10 months ago

你好，这个链接打不开

yzdad commented 10 months ago

You need to convert the GPS to the corresponding tile coordinates. For example, the GPS latitude=48.98254523586602, longitude=8.39036610004500 in the first frame of Kitti's 00 sequence. When converted to the ZOOM=17 tile, it is x=68590.8 y=45022.7, then the corresponding resolution is Pictures available via link: https://gac-geo.googlecnapps.cn/maps/vt?lyrs=s&x=68590&y=45022&z=17

hty1233 commented 9 months ago

your work about Geo-localization with Transformer-based 2D-3D match Network is very useful for me,If there is something I don’t understand in the future, I hope you can give me some advice

------------------ 原始邮件 ------------------ 发件人: "yzdad/D-GLSNet" @.>; 发送时间: 2023年12月12日(星期二) 中午1:49 @.>; @.**@.>; 主题: Re: [yzdad/D-GLSNet] satellite image of kitti-360 (Issue #3)

You need to convert the GPS to the corresponding tile coordinates. For example, the GPS latitude=48.98254523586602, longitude=8.39036610004500 in the first frame of Kitti's 00 sequence. When converted to the ZOOM=17 tile, it is x=68590.8 y=45022.7, then the corresponding resolution is Pictures available via link: https://gac-geo.googlecnapps.cn/maps/vt?lyrs=s&x=68590&y=45022&z=17

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>

hty1233 commented 9 months ago

Hello, can you open source datasets? It's a bit difficult for me to make it myself

yzdad commented 9 months ago

Sorry for the late reply, I've been busy with my thesis recently. To facilitate training, I packaged all the data into XX.npy, with a total of 500 GB. The main problem is that I don't have enough cloud space to upload this data.

hty1233 commented 9 months ago

Thank you for taking the time to reply to my email. Now I have changed my approach to conducting my experiment. If data is needed later, I will contact you again. Once again, I sincerely thank you for your reply.

------------------ 原始邮件 ------------------ 发件人: "yzdad/D-GLSNet" @.>; 发送时间: 2023年12月29日(星期五) 下午3:45 @.>; @.**@.>; 主题: Re: [yzdad/D-GLSNet] satellite image of kitti-360 (Issue #3)

Sorry for the late reply, I've been busy with my thesis recently. To facilitate training, I packaged all the data into XX.npy, with a total of 500 GB. The main problem is that I don't have enough cloud space to upload this data.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>

lbhwyy commented 4 months ago

You need to convert the GPS to the corresponding tile coordinates. For example, the GPS latitude=48.98254523586602, longitude=8.39036610004500 in the first frame of Kitti's 00 sequence. When converted to the ZOOM=17 tile, it is x=68590.8 y=45022.7, then the corresponding resolution is Pictures available via link: https://gac-geo.googlecnapps.cn/maps/vt?lyrs=s&x=68590&y=45022&z=17

Hello author, the download link you provided is currently inaccessible. Do you have an alternative download link available?

yzdad commented 4 months ago

You can try this link: https://mt2.google.com/vt?lyrs=s&x=68590&y=45022&z=17

lbhwyy commented 4 months ago

You can try this link: https://mt2.google.com/vt?lyrs=s&x=68590&y=45022&z=17

It works, thank you.

lbhwyy commented 4 months ago

Sorry for the late reply, I've been busy with my thesis recently. To facilitate training, I packaged all the data into XX.npy, with a total of 500 GB. The main problem is that I don't have enough cloud space to upload this data.

For me, downloading satellite images is too slow, and network errors often occur. Hey, I can provide a Baidu Netdisk VIP. This way, you'll have enough cloud space for dataset uploads. Could you please upload the whole dataset? Thank you.