Download presentation
Presentation is loading. Please wait.
Published byWhitney McCarthy Modified over 9 years ago
1
Multi-Local Feature Manifolds for Object Detection Oscar Danielsson (osda02@csc.kth.se) Stefan Carlsson (stefanc@csc.kth.se) Josephine Sullivan (sullivan@csc.kth.se) DICTA08
2
The Problem Object categories are often modeled by collections (bag-of-features) or constellations (pictorial structures) of local features Many simple, shape-based objects don’t have any discriminative local appearance features ?
3
The Multi-Local Feature A specific spatial constellation of oriented edgels (or other local content) Captures global shape properties “Weak” detector of shape-based object categories Described by coordinate vector: (x 1,…,x 12 )
4
Modeling Intra-Class Variation
5
1. Generate coordinate vectors by clicking corresponding edgels in a (small) number of training images 2. Align coordinate vectors wrt. similarity transform
6
Modeling Intra-Class Variation 3. Extend coordinate vectors into their convex hull
7
Detection
8
For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
9
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
10
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
11
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
12
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
13
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
14
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
15
Detection For each occurrence x 1 of c 1 For each consistent occurrence x 2 of c 2 Sample from p(x 4,x 3 |x 2,x 1 ) to hypothesize image locations of c 3 and c 4 Sample image edgels Compute normalized distance to convex hull of training features If distance is below threshold, an instance was detected End For
16
Experiments Detection performance was evaluated on a standard database (ETHZ Shape Classes) and we want to investigate: Is a multi-local feature a good weak detector? How many local features should be used?
17
Mugs - Training 3 1 8 10 149 7 1213 2 6 11 5 4 3 1 8 10 14 9 7 12 13 2 6 11 5 4 25 training images were downloaded from Google images 14 edgels constituting a multilocal feature were marked in each training image
18
Mugs - Results
19
Performance decreases when adding more than 9 local features 0.4 60.6 %
20
Bottles - Training 12 1 10 7 11 9 8 6 2 5 3 4 1 10 7 11 9 8 6 2 5 3 4 12 25 training images were downloaded from Google images 12 edgels constituting a multilocal feature were marked in each training image
21
Bottles - Results
22
0.4 72.7 %
23
Apple logos - Training 20 training images were downloaded from Google images 12 edgels constituting a multilocal feature were marked in each training image
24
Apple logos - Results
25
Performance decreases when adding more than 11 local features 0.4 77.3 %
26
Conclusions A multi-local feature is a good weak detector of shape-based object categories The best performance is achieved with multi- local features with a moderate number of local features Convex combinations of valid exemplars are in general also valid exemplars (we can extend a few training examples into their convex hull)
27
Future Work Automatic learning of multi-local features Building combinations of multi-local feature detectors into an object detection system
28
Related Work Pictorial Structures E.g.. Felzenszwalb, Huttenlocher. Pictorial Structures for Object Recognition, IJCV No. 1, January 2005. Constellation Models E.g.. Fergus, Perona, Zisserman. Object class recognition by unsupervised scale-invariant learning, CVPR03. Differences Different detection methods Use rich local features
29
Thanks!
30
Representation The multi-local feature manifold consists of all convex combinations of the training examples
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.