search

hehefan / Unsupervised-Person-Re-identification-Clustering-and-Fine-tuning

219 stars 87 forks source link

Similarity matrix #1

Closed ahangchen closed 7 years ago

ahangchen commented 7 years ago

Hi, I'm also implementing person re-id in keras and found this repository, but I'm confused about the similarity matrix computation in baseline evaluation code.

Generally, we compute similarities between probe features and gallery features using euclidean distance formula, so the value in the similarity matrix position(i,j) will be (probe[i] - gallery[j]).^2. But in this code. the value in the similarity matrix is two vectors's dot product: probe[i]*gallery[j].

I wonder whether this metric really works. Is there any references about this metric that compute similarity by vectors' dot product?

ahangchen commented 7 years ago

Got it, it's the cosine similary measure metric.

Mickeybaby commented 5 years ago

Hi，@ahangchen I want to ask you a question. When i run 'python unsupervised.py' FileNotFoundError: [Errno 2] No such file or directory: '../dataset/Duke/bounding_box_train/0001_c2_f0046182.jpg' The bounding_box_train folder is not used in the dataset! Can you give me some advice?Because the author seems to have abandoned the repo. I'll be grateful. Best wishes to you!

ahangchen commented 5 years ago

@Mickeybaby I think bounding_box_train is train directory in the dataset. You can search for 0001_c2_f0046182.jpg to confirm this.