Bencheng

@Rkyzzy Thanks for your interest! VectorMapNet does not use [-51.2, -51.2, -5.0, 51.2, 51.2, 3.0]. The [`point_cloud_range`](https://github.com/Mrmoore98/VectorMapNet_code/blob/4bcc8b4d6e6d61066ff0fd20484d9f5852a7a62f/configs/vectormapnet.py#L9) hyperparameter is just a placeholder. It does not pass `point_cloud_range` to any function....

Hi @tanjatang , thanks for your interest! Currently, MapTR does not support topology prediction. An intuitive way to support topology prediction is adding a topology prediction head following [STSU](https://github.com/ybarancan/STSU). And...

Other people have contacted me about this error. But I rerun my script in my environment and didn't reproduce this error. Can you provide me with your specific environmental setting?...

@ldhai I just upload centerline ckpt & log, please check it out.

> > @ldhai I just upload centerline ckpt & log, please check it out. > > Thank you very much! Hmmmmm, can you update that visualization code, it's the maptrV2...

> > Other people have contacted me about this error. But I rerun my script in my environment and didn't reproduce this error. Can you provide me with your specific...

> i have the same problem. i update networkx from 2.2 to 2.3 and fix this problem Thanks for your help and remind. My networkx version is 2.8.7

You should add extra color for the centerline semantic. We will update the visualization code and document in recent days.

We have not tried this combination. Since NMP focuses on improving BEV feature representation with historical cached feature, which is orthogonal to MapTRv2. We will consider adding this feature in...

You can refer to our TABLE 3. Bevpool can explicitly utilize the depth information and significantly boost the performance.