edits

app.py

@@ -58,7 +58,25 @@ transform = transforms.Compose([
 
 # ---------------------------------------    
 
-
+def match(query_feat, pos_feat, LoweRatioTh=0.9):
+    # first perform reciprocal nn
+    dist = torch.cdist(query_feat, pos_feat)
+    best1 = torch.argmin(dist, dim=1)
+    best2 = torch.argmin(dist, dim=0)
+    arange = torch.arange(best2.size(0), device=best2.device)
+    reciprocal = best1[best2]==arange
+    # check Lowe ratio test
+    dist2 = dist.clone()
+    dist2[best2,arange] = float('Inf')
+    dist2_second2 = torch.argmin(dist2, dim=0)
+    ratio1to2 = dist[best2,arange] / dist2_second2
+    valid = torch.logical_and(reciprocal, ratio1to2<=LoweRatioTh)
+    pindices = torch.where(valid)[0]
+    qindices = best2[pindices]
+    # keep only the ones with same indices 
+    valid = pindices==qindices
+    return pindices[valid]
+    
 
 # sf_idx_ = [55, 14, 5, 4, 52, 57, 40, 9]
 
@@ -97,6 +115,12 @@ def generate_matching_superfeatures(im1, im2, scale_id=6, threshold=50, sf_ids='
         feats2 = output2[0][0]
         attns2 = output2[1][0]
         strenghts2 = output2[2][0]
+
+        feats1n = F.normalize(feats1, dim=1)
+        feats2n = F.normalize(feats2, dim=1)
+        ind_match = match(feats1n, feats2n)
+        print('ind', ind_match)
+        print('ind.shape', ind_match.shape)
         # outputs = []
         # for im_tensor in loader:
         #     outputs.append(net.get_superfeatures(im_tensor.to(device), scales=[scales[scale_id]]))