How to mount the data repository
The data repository can be found here or mounted as descibed below.
There is no password or username necessary for read-only access.
Simply press “Map network drive” from the Windows Explorer and enter the link provided above.
sudo mount -t davfs https://segem.rzg.mpg.de/webdav SOME_MOUNTPOINT (make sure you have the related packages)
Brief explanantion of content
The content is located in three main folders:
- onlineMaterial, online material as described in the paper
- datasets - The dataset ek_0563 (raw data, classification and segmentation)
- supplement, all supplements to the paper
- SegEM_challenge, dense skeleton data (and corresponding raw data) for two regions from S1 L4 and the volume annotated training regions
Detailed explanation of content:
Supplemental Data (/supplement)
- Gallery of retina volume training data (Data S1.pdf).
- All trained CNNs used in main figures (Data S2.zip)
- All dense skeleton data (Data S3.zip)
- Gallery of all skeleton-to-segmentation object overlays (retina test; at node threshold 1; for segmentation with parameters also shown in Fig. 3). Shown volumes are of size 4.6 * 4.6 * 9.6 μm 3 . (Data S4.pdf)
- Gallery of all mergers detected in retina test volume (Data S5.pdf)
- Whole cell skeleton reconstructions from the mouse retina inner plexiform layer as used in Figure 4 (Data S6.zip)
- Whole cell skeleton reconstructions from mouse barrel cortex layer 4 as seen in Figure 5 (Data S7.zip)
- SegEM code (Supplemental Code, Data S8.zip, see segem.io for continuously updated versions, compare to workflow in Figure 1)
- Movies of top scoring (with respect to rand and warp error) test stack classifications as used for ISBI metric calculation (Data S9.zip).
Online Material (/onlineMaterial)
- All volume training data, retina dataset (retinaTrainingData.zip)
- CNN classifiers trained on retina dataset (retinaTrainedCNNs.zip)
- All volume training data, cortex dataset (cortexTrainingData.zip)
- CNN classifiers trained on cortex dataset (cortexTrainedCNNs.zip)
- Raw EM-data dataset from mouse retina (/ek0563/raw)
- Classification of the dataset (/ek0563/class)
- Segmentation of the dataset (/ek0563/seg)
SegEM challenge (/SegEM_challenge)
- Densely skeltonized training and test regions as .nml (/skeletonData/cortex_training.nml and cortex_test.nml)
- Raw data (of densly annotated regions including border for classifier pixel field of view, /skeletonData/cortex_training_raw.mat and cortex_test_raw.mat)
- Densely volume annotated regions for classifier training & testing (/volumeTrainingData)