- QTnet
- Creating a difficult testing datasets. And make a comparison about the model on the easy testing datasets and difficult testing datasets.
- location: mask; combination: mask; fusion;
- easy: 0.897; 0.987; 0.987; difficult: 0.816; 0.931; 0.909
- location: mask; combination: mask;
- easy: 0.883; 0.964; 0.961; difficult: 0.816; 0.932; 0.885
- location: bbox; combination: mask; fusion;
- easy: 0.878; 0.979; 0.979; … Continue Reading ››
- easy: 0.878; 0.979; 0.979; … Continue Reading ››
- location: mask; combination: mask; fusion;
- Creating a difficult testing datasets. And make a comparison about the model on the easy testing datasets and difficult testing datasets.
zhou のすべての投稿
先週の進捗(ZHOU)
- QT-Net:
- Add the result on the test datasets to the report (git), the requirement tools (requirement)
- location: mask; combination: mask; fusion; 0.897; 0.987; 0.987
- location: mask; combination: mask; 0.883; 0.964; 0.961
- location: bbox; combination: mask; fusion; 0.878; 0.979; 0.979
- location: bbox; combination: mask; 0.891; 0.971; … Continue Reading ››
先週の進捗(ZHOU)
Finished the testing work, the result shows that more image is used to train lead better performance on testing datasets. (git)
先週の進捗(ZHOU)
- QT-Net:
- Taken some photos for testing the network;
- Finished one experiment about the influence of illumination.
- I will update the experiment report including the analysis about the performance on test datasets.
先週の進捗(ZHOU)
- QT-NET: Continue to do experiment about the number of image in each illumination. But the experiment doesn't finished and this may take another month.
- SLAM: About the experiments result, I make an analysis. But, there are some problems that I need to rethink.
先週の進捗(ZHOU)
- QT-NET (git)
- Finished the report of experiment 6 about illumination.
- Make an experiment plan about the number of images in each illumination, and do a part of the experiments.
- SLAM
- Two methods are used train the network: (1) all pixels, (2) consistent pixels.
- The pose estimation can be improved with the combination of (1) and (2). … Continue Reading ››
先週の進捗(ZHOU)
- QT-NET
- Finished the experiments about the illumination, include 9, 5, 3, 1 kinds of illumination. The result is upload to git.
- SLAM
- continue to do some experiments about the consistency of pixel.
先週の進捗(ZHOU)
- QT-NET
- Finished the 1.0 version of the research report, and upload to the git. But, some grammar error and express error need to be modified in the future.
- The experiment about 3 group of illumination is finished today, next week I will try to test the network with 1 group of illumination. Due … Continue Reading ››
先週の進捗(ZHOU)
- About the positive and negative group, I do some experiments to test the influence of illumination. For 9, 5, 3 group of illumination, I compared the accuracy and loss. And all the experiment is upload to git.
- The result is that maybe illumination just has a litter influence to the accuracy.
Continue Reading ››
先週の進捗(ZHOU)
- QT-Net
- About the positive and negative group, I do some experiments to test the influence of times of each object are used for combination.
- SLAM
- I am doing some work to add the occlusion mask, consistence mask, and consistence score … Continue Reading ››
- I am doing some work to add the occlusion mask, consistence mask, and consistence score … Continue Reading ››