Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit7d79106

Browse files
committed
Fix model and log links and update README
1 parent877a2bd commit7d79106

File tree

1 file changed

+31
-21
lines changed

1 file changed

+31
-21
lines changed

‎README.md‎

Lines changed: 31 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -210,7 +210,7 @@ Finally, to pack the prediction file into the submission format, please modify t
210210
python tools/submit_results.py
211211
```
212212

213-
Then you can submit the resulting pkl file to the test server(to go live by the end of March)and wait for the lottery :)
213+
Then you can submit the resulting pkl file to the test server and wait for the lottery :)
214214

215215
We also provide a sample script`tools/eval_script.py` for evaluating the submission file and you can check it by yourself to ensure your submitted file has the correct format.
216216

@@ -224,45 +224,55 @@ Note that the performance is a little different from the results provided in the
224224

225225
| Method| Input| AP@0.25| AR@0.25| AP@0.5| AR@0.5| Download|
226226
|:------:|:-----:|:-------:|:-------:|:------:|:------:|:------:|
227-
|[Baseline](configs/detection/mv-det3d_8xb4_embodiedscan-3d-284class-9dof.py)| RGB-D| 15.22| 52.23| 8.13| 26.66|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-3ddet.pth&sign=463b8f3d1018a2308933c4880eb45809&nonce=1713253422165),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-3ddet.log&sign=31e1b4d08fcdc1b849ad997d3986355c&nonce=1713253411612)|
227+
|[Baseline](configs/detection/mv-det3d_8xb4_embodiedscan-3d-284class-9dof.py)| RGB-D| 15.22| 52.23| 8.13| 26.66|[Model](https://download.openmmlab.com/mim-example/embodiedscan/mv-3ddet.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-3ddet.log)|
228228

229229
####Continuous 3D Detection
230230

231231
| Method| Input| AP@0.25| AR@0.25| AP@0.5| AR@0.5| Download|
232232
|:------:|:-----:|:-------:|:-------:|:------:|:------:|:------:|
233-
|[Baseline](configs/detection/cont-det3d_8xb1_embodiedscan-3d-284class-9dof.py)| RGB-D| 17.83| 47.53| 9.04| 23.04|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=cont-3ddet.pth&sign=33a9d8c7269d9cb8a6c12596eccf33fb&nonce=1713253437312),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=cont-3ddet.log&sign=1941012e5c4de98852fe64d4c7ca8c2b&nonce=1713253431164)|
234-
235-
####Multi-View Occupancy Prediction
236-
237-
| Method| Input| mIoU| Download|
238-
|:------:|:-----:|:----:|:--------:|
239-
|[Baseline](configs/occupancy/mv-occ_8xb1_embodiedscan-occ-80class.py)| RGB-D| 21.28|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-occ.pth&sign=994480cb17c8522c963a5ac8df699a93&nonce=1713253452434),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-occ.log&sign=edca607abc7ed35f369dda0552caf732&nonce=1713253459552)|
240-
241-
####Continuous Occupancy Prediction
242-
243-
| Method| Input| mIoU| Download|
244-
|:------:|:-----:|:----:|:--------:|
245-
|[Baseline](configs/occupancy/cont-occ_8xb1_embodiedscan-occ-80class.py)| RGB-D| 22.92|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=cont-occ.pth&sign=c561946827abbae17a260cb85765bfc5&nonce=1713253478620),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=cont-occ.log&sign=06c5dd32250079257b1469d4e2b34a93&nonce=1713253472602)|
233+
|[Baseline](configs/detection/cont-det3d_8xb1_embodiedscan-3d-284class-9dof.py)| RGB-D| 17.83| 47.53| 9.04| 23.04|[Model](https://download.openmmlab.com/mim-example/embodiedscan/cont-3ddet.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/cont-3ddet.log)|
246234

247235
####Multi-View 3D Visual Grounding
248236

249237
| Method|AP@0.25| AP@0.5| Download|
250238
|:------:|:-----:|:-------:|:------:|
251-
|[Baseline-Mini](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof.py)| 33.59| 14.40|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding.pth&sign=b88b76044eadb27aa035a30390d358f8&nonce=1713253489268),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding.log&sign=5fb591b76407097792a3a2bdb885fa45&nonce=1713253496042)|
239+
|[Baseline-Mini](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof.py)| 33.59| 14.40|[Model](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding.log)|
252240
|[Baseline-Mini (w/ FCAF box coder)](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof_fcaf-coder.py)| -| -| -|
253-
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| 36.78| 15.97|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-full.pth&sign=6e0ecfb03f3cf9bf248c9b0a1692dd19&nonce=1713253505107),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-full.log&sign=b658bb9b731e3cf4351d3aac26ff02ed&nonce=1713253511050)|
241+
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| 36.78| 15.97|[Model](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-full.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-full.log)|
254242

255243
Note: As mentioned in the paper, due to much more instances annotated with our new tools and pipelines, we concatenate several simple prompts as more complex ones to ensure those prompts to be more accurate without potential ambiguity. The above table is the benchmark without complex prompts using the initial version of visual grounding data.
256244

257245
We found such data is much less than the main part though, it can boost the multi-modal model's performance a lot. Meanwhile, whether to include these data in the validation set is not much important. We provide the updated benchmark as below and update a version of visual grounding data via emails to the community.
258246

259247
| Method| train| val| AP@0.25| AP@0.5| Download|
260248
|:------:|:-----:|:---:|:------:|:-----:|:--------:|
261-
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| w/o complex| w/o complex| 36.78| 15.97|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-full.pth&sign=6e0ecfb03f3cf9bf248c9b0a1692dd19&nonce=1713253505107),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-full.log&sign=b658bb9b731e3cf4351d3aac26ff02ed&nonce=1713253511050)|
262-
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| w/ complex| w/o complex| 39.26| 18.86|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-complex.pth&sign=fe1fb560a3d51f59486c353d5748081f&nonce=1713253535423),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-complex.log&sign=af7c5dcb19cadc40cc64c15b8c5d5d51&nonce=1713253542450)|
263-
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| w/ complex| w/ complex| 39.21| 18.84|[Model](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-complex.pth&sign=fe1fb560a3d51f59486c353d5748081f&nonce=1713253535423),[Log](https://download.openxlab.org.cn/repos/file/wangtai/EmbodiedScan/main?filepath=mv-grounding-complex.log&sign=af7c5dcb19cadc40cc64c15b8c5d5d51&nonce=1713253542450)|
249+
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| w/o complex| w/o complex| 36.78| 15.97|[Model](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-full.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-full.log)|
250+
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| w/ complex| w/o complex| 39.26| 18.86|[Model](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-complex.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-complex.log)|
251+
|[Baseline-Full](configs/grounding/mv-grounding_8xb12_embodiedscan-vg-9dof-full.py)| w/ complex| w/ complex| 39.21| 18.84|[Model](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-complex.pth),[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-grounding-complex.log)|
252+
253+
####Multi-View Occupancy Prediction
254+
255+
| Method| Input| mIoU| Download|
256+
|:------:|:-----:|:----:|:--------:|
257+
|[Baseline](configs/occupancy/mv-occ_8xb1_embodiedscan-occ-80class.py)| RGB-D| 21.28|[Log](https://download.openmmlab.com/mim-example/embodiedscan/mv-occ.log)|
258+
259+
####Continuous Occupancy Prediction
260+
261+
| Method| Input| mIoU| Download|
262+
|:------:|:-----:|:----:|:--------:|
263+
|[Baseline](configs/occupancy/cont-occ_8xb1_embodiedscan-occ-80class.py)| RGB-D| 22.92|[Log](https://download.openmmlab.com/mim-example/embodiedscan/cont-occ.log)|
264+
265+
Because the occupancy prediction models are a little large, we save them via OpenXLab and do not provide direct download links here. To download these checkpoints on OpenXLab, please run the following commands:
266+
267+
```bash
268+
# If you did not install LFS before
269+
git lfs install
270+
# git clone EmbodiedScan model repo via
271+
git clone https://code.openxlab.org.cn/wangtai/EmbodiedScan.git
272+
# Then you can cd EmbodiedScan to get all the pretrained models
273+
```
264274

265-
Please see the[paper](./assets/EmbodiedScan.pdf) for more details of ourtwo benchmarks, fundamental 3D perception and language-groundedbenchmarks. This dataset is still scaling up and the benchmark is being polished and extended. Please stay tuned for our recent updates.
275+
Please see the[paper](./assets/EmbodiedScan.pdf) for more details of our benchmarks. This dataset is still scaling up and the benchmark is being polished and extended. Please stay tuned for our recent updates.
266276

267277
##📝 TODO List
268278

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp