Eli-YiLi

Training time varies from different depth. Ｍulti-branch in Pytorch can not be processed in parallel, the training time is little slower than ResNets. TensorFlow ｒｕｎｓ faster.

Use layer[1-3] in scalenet.py and add a json file to describe the structure in the paper supplement.

简单来说逐block逐channel的对通道进行评分，参照标准是bn的scale参数，取预定计算复杂度下top的bn_scale对应的神经元并统计所在尺度。评分除了依照scale还要考虑计算复杂度，scale/cost，论文中为简化超参，cost加了一个0次指数(只看scale)。另外避免如剪枝一样在全局优化只对当前block进行排序，而非全局。 在 2019-06-24 11:51:03，"GuideWsp" 写道： 我想了解一下如何将一个多分支相同通道数的网络按照你论文中算法1裁剪为多分支不同通道数的网络。 我认为该算法才是ScalaNet的核心。 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

> I have been working on segmenting small objects(based on patch size value) with low-quality image descriptions, such as car blinkers. I have used a sliding window technique to extract...

I come again ... It's an excellent work above via grounding box. While we provide a simpler solution via CLIP's explainability. Our work can achieve text to mask with SAM...

Our work can achieve text to mask with SAM: https://github.com/xmed-lab/CLIP_Surgery This is our work about CLIP's explainability. It's able to guide SAM to achieve text to mask without manual points....