Aldo Zaimi
Aldo Zaimi
ref: https://github.com/neuropoly/axondeepseg/pull/52/files/286dc698db156e71a7271a25016a442e8e2de450#r170419632
When segmenting a large scale sample (e.g. full spinal cord slice), users need to obtain downsampled morphometrics maps. This can currently be done by the AxonSeg pipeline. The implementation of...
So far, I experimented with distance transform + watershed, but it is only robust on samples with low axon density (less compact fibers) and/or neighbour myelin sheaths that have a...
- if holes in axons or myelin - only needed if model is not good enough
- we currently use 0.75 - papers suggest 0.5 for hidden units and 0.1-0.3 for input layer
Hello, I had 2 questions regarding the license (mostly about the MobileViT model): - Does the Apple license allow commercial use for models attached to it? - What about the...
Hello there, I was wondering about the license of the pretrained weights of this model (for example the ones on HuggingFace, such as https://huggingface.co/apple/deeplabv3-mobilevit-small). The code has an MIT license,...
Hello, I was wondering what is the state of the license of SegFormer pretrained weights available in other platforms such as HuggingFace (also related to this discussion: https://github.com/NVlabs/SegFormer/issues/20). On HuggingFace,...
Hello, I am trying to find a way to compute the CLIP directional similarity on my original vs edited samples, but I do not have captions for my samples. I...
Hey there, I am interested in running VQAScore with another VLM, CogVLM (see [here](https://huggingface.co/THUDM/cogvlm-chat-hf)). I was looking at the guidelines on how to adapt to another VQA model checking the...