Object Classification Using Heterogeneous Co-occurrence Features

Ito, Satoshi; Kubota, Susumu

doi:10.1007/978-3-642-15555-0_51

Cited by 18 publications

(22 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of our approach on this dataset (see Table 1) is 80.66%, which outperforms all previous known methods in the literature (some by as much as 4 to 8%) [3,12,17,18]. One important thing to note is that the improvement of our algorithm over our baseline is about 4%, and the only difference between the two is the addition of the proposed segmentation algorithm and the features extracted from the segmented image.…”

Section: Oxford 102 Flower Species Datasetmentioning

confidence: 66%

“…Accuracy (in %) Our baseline (no segmentation) 76.7 Nilsback and Zisserman [17] 72.8 Ito and Cubota [12] 74.8 Nilsback and Zisserman [18] 76.3 Chai, Bicos method [3] 79.4 Chai, BicosMT method [3] 80.0 Ours 80.66 Ours: improvement over our baseline +3.94 Table 1. Classification performance on Oxford 102 flower dataset.…”

Section: Methodsmentioning

confidence: 99%

“…A lot of methods have been tested on this dataset [3,12,17,18], including some segmentation-based [3,17]. Some of the segmentation methods are designed to be very specific to the appearance of flowers [17] (with the assumption that a single flower is in the center of the image and takes most of the image), while others [3] are more general and can also be applied to other types of datasets.…”

Section: Oxford 102 Flower Species Datasetmentioning

confidence: 99%

See 2 more Smart Citations

Efficient Object Detection and Segmentation for Fine-Grained Recognition

Angelova¹,

Zhu²

2013

2013 IEEE Conference on Computer Vision and Pattern Recognition

199

129

View full text Add to dashboard Cite

We propose a detection and segmentation algorithm for the purposes of fine-grained recognition. The algorithm first detects low-level regions that could potentially belong to the object and then performs a full-object segmentation through propagation. Apart from segmenting the object, we can also 'zoom in' on the object, i.e. center it, normalize it for scale, and thus discount the effects of the background. We then show that combining this with a state-of-the-art classification algorithm leads to significant improvements in performance especially for datasets which are considered particularly hard for recognition, e.g. birds species.The proposed algorithm is much more efficient than other known methods in similar scenarios [4,21]. Our method is also simpler and we apply it here to different classes of objects, e.g. birds, flowers, cats and dogs.We tested the algorithm on a number of benchmark datasets for fine-grained categorization. It outperforms all the known state-of-the-art methods on these datasets, sometimes by as much as 11%. It improves the performance of our baseline algorithm by 3-4%, consistently on all datasets. We also observed more than a 4% improvement in the recognition performance on a challenging largescale flower dataset, containing 578 species of flowers and 250,000 images.

show abstract

Section: Oxford 102 Flower Species Datasetmentioning

confidence: 66%

Section: Methodsmentioning

confidence: 99%

Section: Oxford 102 Flower Species Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Object Detection and Segmentation for Fine-Grained Recognition

Angelova¹,

Zhu²

2013

2013 IEEE Conference on Computer Vision and Pattern Recognition

199

129

View full text Add to dashboard Cite

show abstract

“…Although the extracted features, such as color, texture, shape, or motion features, can be quite weak individually, an appropriate combination of them will bring a strong feature which is much more discriminative [31] [29] [3] [16] [2] [9]. There has been a recent trend in mining co-occurrence patterns for visual recognition.…”

Section: Introductionmentioning

confidence: 99%

“…In spite of many previous works in mining and integrating co-occurrence patterns [22] [20] [3] [29] [33] [16] [24] [9] [27], none of these methods is targeted at finding the most discriminative co-occurrence pattern with the smallest classification error. Given N binary features, because the co-occurrence pattern can contain an arbitrary number of features (up to N ), the total number of candidates of co-occurrence patterns is exponentially large (e.g.…”

Section: Introductionmentioning

confidence: 99%