Comments (13)
@deepmeng, could you share evaluation results here? As far as I know, there are no available KITTI evaluation results for this method. Thanks
from sfa3d.
Hi @deepmeng
Yeb, this work has a better performance compared to the YOLO-based methods in both accuracy and speed.
Please contribute the full evaluation tool if you make it. Thanks.
from sfa3d.
Hi @deepmeng and @maudzung
I couldn't figure out how I can evaluate (write the prediction results as txt files). Could you share your evaluation file.
Thanks
from sfa3d.
@bilgecanpullinen OK. The code in my hand is chaotic. I will fine-tune and share the code.
from sfa3d.
Hi @iliiliiliili and @bilgecanpullinen . I forked this repo and I added the evaluation tools. I didn't merge my code to this repo because my code is not concise. You can refer to this repo (https://github.com/reinforcementdriving/Super-Fast-Accurate-3D-Object-Detection), extract and adapt the related code to evaluate the prediction results.
from sfa3d.
from sfa3d.
Hi @iliiliiliili and @bilgecanpullinen . I forked this repo and I added the evaluation tools. I didn't merge my code to this repo because my code is not concise. You can refer to this repo (https://github.com/reinforcementdriving/Super-Fast-Accurate-3D-Object-Detection), extract and adapt the related code to evaluate the prediction results.
Thank you, @deepmeng. Could you add evaluation results on KITTI validation subset with your code to README? It would be very helpful 🙂
from sfa3d.
@bilgecanpullinen
Please try to modify the function 'create_val_dataloader(configs)' at line 54 in 'sfa/data_process/kitti_dataloader.py':
'return val_dataloader, val_dataset'.
Thank you I have tried but now I have some problem with the dataloader.; eval.py line 112 val_dataloader, val_dataset = create_val_dataloader(configs) gives me a value error, which is too many values to unpack(expected 2). But you have used at your train.py the same way for create_train_dataloader. So do you have any idea how I can manage this one. deepmeng @.***>, 30 Mar 2021 Sal, 05:51 tarihinde şunu yazdı:
…
Hi @iliiliiliili https://github.com/iliiliiliili and @bilgecanpullinen https://github.com/bilgecanpullinen . I forked this repo and I added the evaluation tools. I didn't merge my code to this repo because my code is not concise. You can refer to this repo ( https://github.com/reinforcementdriving/Super-Fast-Accurate-3D-Object-Detection), extract and adapt the related code to evaluate the prediction results. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUM3GN4XDV5TBCYOF2OFDTTGE4C5ANCNFSM4TOLZ7LQ .
from sfa3d.
from sfa3d.
Thank you again. Everything worked good until the kitti-eval now. Sorry for asking again to you but it seems you are using the same kitti-eval-python tool. I am receiving ERROR: The function received no value for the required argument: label_split_file Usage: evaluate.py evaluate LABEL_PATH RESULT_PATH LABEL_SPLIT_FILE optional flags: --current_class | --coco | --score_thresh For detailed information on this command, run: evaluate.py evaluate --help which is from the label_split_file and I am sure that my path is correct, I tried with from https://xiaozhichen.github.io/files/mv3d/imagesets.tar.gz file as well. Do you have any suggestions. deepmeng @.>, 30 Mar 2021 Sal, 12:10 tarihinde şunu yazdı:
…
@bilgecanpullinen https://github.com/bilgecanpullinen Please try to modify the function 'create_val_dataloader(configs)' at line 54 in 'sfa/data_process/kitti_dataloader.py': 'return val_dataloader, val_dataset'. Thank you I have tried but now I have some problem with the dataloader.; eval.py line 112 val_dataloader, val_dataset = create_val_dataloader(configs) gives me a value error, which is too many values to unpack(expected 2). But you have used at your train.py the same way for create_train_dataloader. So do you have any idea how I can manage this one. deepmeng @.>, 30 Mar 2021 Sal, 05:51 tarihinde şunu yazdı: … <#m_2157938398064785683_> Hi @iliiliiliili https://github.com/iliiliiliili https://github.com/iliiliiliili and @bilgecanpullinen https://github.com/bilgecanpullinen https://github.com/bilgecanpullinen . I forked this repo and I added the evaluation tools. I didn't merge my code to this repo because my code is not concise. You can refer to this repo ( https://github.com/reinforcementdriving/Super-Fast-Accurate-3D-Object-Detection), extract and adapt the related code to evaluate the prediction results. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#16 (comment) <#16 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUM3GN4XDV5TBCYOF2OFDTTGE4C5ANCNFSM4TOLZ7LQ . — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUM3GJUC6FNGKAY7U63JN3TGGIRJANCNFSM4TOLZ7LQ .
I used this tool (https://github.com/prclibo/kitti_eval). If you can't use the kitti eval tool properly, I have no idea.
from sfa3d.
from sfa3d.
Thank you again. I am having another problem and sorry for asking it again but I have tried many things. g++ -o evaluate_object_3d_offline evaluate_object_3d_offline.cpp -O3 evaluate_object_3d_offline.cpp:12:10: fatal error: boost/numeric/ublas/matrix.hpp: No such file or directory #include <boost/numeric/ublas/matrix.hpp> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ compilation terminated. This is my error while I try to compile. Do you remember having an issue like this. deepmeng @.***>, 30 Mar 2021 Sal, 15:07 tarihinde şunu yazdı:
…
Thank you again. Everything worked good until the kitti-eval now. Sorry for asking again to you but it seems you are using the same kitti-eval-python tool. I am receiving ERROR: The function received no value for the required argument: label_split_file Usage: evaluate.py evaluate LABEL_PATH RESULT_PATH LABEL_SPLIT_FILE optional flags: --current_class | --coco | --score_thresh For detailed information on this command, run: evaluate.py evaluate --help which is from the label_split_file and I am sure that my path is correct, I tried with from https://xiaozhichen.github.io/files/mv3d/imagesets.tar.gz file as well. Do you have any suggestions. deepmeng @. >, 30 Mar 2021 Sal, 12:10 tarihinde şunu yazdı: … <#m_6312827631054034898_> @bilgecanpullinen https://github.com/bilgecanpullinen https://github.com/bilgecanpullinen https://github.com/bilgecanpullinen Please try to modify the function 'create_val_dataloader(configs)' at line 54 in 'sfa/data_process/kitti_dataloader.py': 'return val_dataloader, val_dataset'. Thank you I have tried but now I have some problem with the dataloader.; eval.py line 112 val_dataloader, val_dataset = create_val_dataloader(configs) gives me a value error, which is too many values to unpack(expected 2). But you have used at your train.py the same way for create_train_dataloader. So do you have any idea how I can manage this one. deepmeng @.>, 30 Mar 2021 Sal, 05:51 tarihinde şunu yazdı: … <#m_2157938398064785683_> Hi @iliiliiliili https://github.com/iliiliiliili https://github.com/iliiliiliili https://github.com/iliiliiliili and @bilgecanpullinen https://github.com/bilgecanpullinen https://github.com/bilgecanpullinen https://github.com/bilgecanpullinen . I forked this repo and I added the evaluation tools. I didn't merge my code to this repo because my code is not concise. You can refer to this repo ( https://github.com/reinforcementdriving/Super-Fast-Accurate-3D-Object-Detection), extract and adapt the related code to evaluate the prediction results. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#16 <#16> (comment) <#16 (comment) <#16 (comment)>>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUM3GN4XDV5TBCYOF2OFDTTGE4C5ANCNFSM4TOLZ7LQ . — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#16 (comment) <#16 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUM3GJUC6FNGKAY7U63JN3TGGIRJANCNFSM4TOLZ7LQ . I used this tool (https://github.com/prclibo/kitti_eval). If you can't use the kitti eval tool properly, I have no idea. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUM3GJEVPQ3FCGGZ36P453TGG5H3ANCNFSM4TOLZ7LQ .
Well. I'm not sure your experimental environment. You must make sure you have installed the Boost library. Maybe you can try 'sudo apt-get install libboost-all-dev'.
from sfa3d.
from sfa3d.
Related Issues (20)
- Can you tell me more about what 3D-Object-Detection Architecture that you have used and what part in this git? I am still not clear when reading your README file HOT 2
- Cannot get detections on VLP 16 lidar data HOT 1
- 2D pcd file testing HOT 3
- core ml HOT 1
- corners_int is not int type HOT 1
- Data Augmentation
- what is mean by hm_name, hm_mc, hm_out?
- qt.qpa.xcb: could not connect to display
- [Solved] Problem when running it with GPU
- Train without intensity
- TypeError: Descriptors cannot not be created directly
- Train on my datasets, but performance is poor. HOT 2
- classnames.txt file? HOT 1
- ros node /image_rects no launch
- Pointcloud doesn't fit in the image
- How to measure the precision and accuracy of the model ?
- Data Augmentation Issue in Random Scaling
- Pitch and roll prediction feature request HOT 3
- can we make the algorithm such a way so that it only takes point cloud for training and not images
- what is lidar_aug in kiitti_dataset.py in KittiDataset class?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sfa3d.