Contextual Classification of Image Patches with Latent Aspect Models
We present a novel approach for contextual classification of image patches in complex visual scenes, based on the use of histograms of quantized features and probabilistic aspect models. Our approach uses context in two ways: (1) by using the fact that specific learned aspects correlate with the sem...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2009-02-01
|
| Series: | EURASIP Journal on Image and Video Processing |
| Online Access: | http://dx.doi.org/10.1155/2009/602920 |
| _version_ | 1819087080705228800 |
|---|---|
| author | Florent Monay Pedro Quelhas Jean-Marc Odobez Daniel Gatica-Perez |
| author_facet | Florent Monay Pedro Quelhas Jean-Marc Odobez Daniel Gatica-Perez |
| author_sort | Florent Monay |
| collection | DOAJ |
| description | We present a novel approach for contextual classification of image patches in complex visual scenes, based on the use of histograms of quantized features and probabilistic aspect models. Our approach uses context in two ways: (1) by using the fact that specific learned aspects correlate with the semantic classes, which resolves some cases of visual polysemy often present in patch-based representations, and (2) by formalizing the notion that scene context is image-specific—what an individual patch represents depends on what the rest of the patches in the same image are. We demonstrate the validity of our approach on a man-made versus natural patch classification problem. Experiments on an image collection of complex scenes show that the proposed approach improves region discrimination, producing satisfactory results and outperforming two noncontextual methods. Furthermore, we also show that co-occurrence and traditional (Markov random field) spatial contextual information can be conveniently integrated for further improved patch classification. |
| first_indexed | 2024-12-21T21:30:28Z |
| format | Article |
| id | doaj.art-d5b83aafc63e43a1bc8b64a0dd7a74db |
| institution | Directory Open Access Journal |
| issn | 1687-5176 1687-5281 |
| language | English |
| last_indexed | 2024-12-21T21:30:28Z |
| publishDate | 2009-02-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EURASIP Journal on Image and Video Processing |
| spelling | doaj.art-d5b83aafc63e43a1bc8b64a0dd7a74db2022-12-21T18:49:39ZengSpringerOpenEURASIP Journal on Image and Video Processing1687-51761687-52812009-02-01200910.1155/2009/602920Contextual Classification of Image Patches with Latent Aspect ModelsFlorent MonayPedro QuelhasJean-Marc OdobezDaniel Gatica-PerezWe present a novel approach for contextual classification of image patches in complex visual scenes, based on the use of histograms of quantized features and probabilistic aspect models. Our approach uses context in two ways: (1) by using the fact that specific learned aspects correlate with the semantic classes, which resolves some cases of visual polysemy often present in patch-based representations, and (2) by formalizing the notion that scene context is image-specific—what an individual patch represents depends on what the rest of the patches in the same image are. We demonstrate the validity of our approach on a man-made versus natural patch classification problem. Experiments on an image collection of complex scenes show that the proposed approach improves region discrimination, producing satisfactory results and outperforming two noncontextual methods. Furthermore, we also show that co-occurrence and traditional (Markov random field) spatial contextual information can be conveniently integrated for further improved patch classification.http://dx.doi.org/10.1155/2009/602920 |
| spellingShingle | Florent Monay Pedro Quelhas Jean-Marc Odobez Daniel Gatica-Perez Contextual Classification of Image Patches with Latent Aspect Models EURASIP Journal on Image and Video Processing |
| title | Contextual Classification of Image Patches with Latent Aspect Models |
| title_full | Contextual Classification of Image Patches with Latent Aspect Models |
| title_fullStr | Contextual Classification of Image Patches with Latent Aspect Models |
| title_full_unstemmed | Contextual Classification of Image Patches with Latent Aspect Models |
| title_short | Contextual Classification of Image Patches with Latent Aspect Models |
| title_sort | contextual classification of image patches with latent aspect models |
| url | http://dx.doi.org/10.1155/2009/602920 |
| work_keys_str_mv | AT florentmonay contextualclassificationofimagepatcheswithlatentaspectmodels AT pedroquelhas contextualclassificationofimagepatcheswithlatentaspectmodels AT jeanmarcodobez contextualclassificationofimagepatcheswithlatentaspectmodels AT danielgaticaperez contextualclassificationofimagepatcheswithlatentaspectmodels |