Julia Astegiano - Academia.edu (original) (raw)

Uploads

Papers by Julia Astegiano

Object tracking is a long standing problem in vision. While great efforts have been spent to impr... more Object tracking is a long standing problem in vision. While great efforts have been spent to improve tracking performance, a simple yet reliable prior knowledge is left unexploited: the target object in tracking must be an object other than non-object. The recently proposed and popularized objectness measure provides a natural way to model such prior in visual tracking. Thus motivated, in this paper we propose to adapt objectness for visual object tracking. Instead of directly applying an existing objectness measure that is generic and handles various objects and environments, we adapt it to be compatible to the specific tracking sequence and object. More specifically, we use the newly proposed BING [7] objectness as the base, and then train an object-adaptive objectness for each tracking task. The training is implemented by using an adaptive support vector machine that integrates information from the specific tracking target into the BING measure. We emphasize that the benefit of the proposed adaptive objectness, named ADOB-ING, is generic. To show this, we combine ADOBING with seven top performed trackers in recent evaluations. We run the ADOBING-enhanced trackers with their base trackers on two popular benchmarks, the CVPR2013 benchmark (50 sequences) and the Princeton Tracking Benchmark (100 sequences). On both benchmarks, our methods not only consistently improve the base trackers, but also achieve the best known performances. Noting that the way we integrate objectness in visual tracking is generic and straightforward, we expect even more improvement by using tracker-specific objectness.

Acta Oecologica, 2013

This article appeared in a journal published by Elsevier. The attached copy is furnished to the a... more This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Royal Society Open Science, 2019

Women underrepresentation in science has frequently been associated with women being less product... more Women underrepresentation in science has frequently been associated with women being less productive than men (i.e. the gender productivity gap), which may be explained by women having lower success rates, producing science of lower impact and/or suffering gender bias. By performing global meta-analyses, we show that there is a gender productivity gap mostly supported by a larger scientific production ascribed to men. However, women and men show similar success rates when the researchers' work is directly evaluated (i.e. publishing articles). Men's success rate is higher only in productivity proxies involving peer recognition (e.g. evaluation committees, academic positions). Men's articles showed a tendency to have higher global impact but only if studies include self-citations. We detected gender bias against women in research fields where women are underrepresented (i.e. those different from Psychology). Historical numerical unbalance, socio-psychological aspects and c...

Advances in Ecological Research, 2013

PloS one, 2015

Most flowering plants depend on pollinators to reproduce. Thus, evaluating the robustness of plan... more Most flowering plants depend on pollinators to reproduce. Thus, evaluating the robustness of plant-pollinator assemblages to species loss is a major concern. How species interaction patterns are related to species sensitivity to partner loss may influence the robustness of plant-pollinator assemblages. In plants, both reproductive dependence on pollinators (breeding system) and dispersal ability may modulate plant sensitivity to pollinator loss. For instance, species with strong dependence (e.g. dioecious species) and low dispersal (e.g. seeds dispersed by gravity) may be the most sensitive to pollinator loss. We compared the interaction patterns of plants differing in dependence on pollinators and dispersal ability in a meta-dataset comprising 192 plant species from 13 plant-pollinator networks. In addition, network robustness was compared under different scenarios representing sequences of plant extinctions associated with plant sensitivity to pollinator loss. Species with differe...

Organisms Diversity & Evolution, 2009

Twenty-five biotically pollinated plants of the Chaco Serrano Forest (Có rdoba, Argentina) were s... more Twenty-five biotically pollinated plants of the Chaco Serrano Forest (Có rdoba, Argentina) were studied in order to analyze whether 'flower functionality' is related to the relationship between pollen size and pistil length. Because flower functionality may act on the respective mean values of pollen size and pistil length rather than on intraspecific variation in these traits, we expected (1) a high positive correlation between pollen size and pistil length in a set of sympatric species, independent of their degree of pollination specialization or generalization; and (2) no interspecific correlation between the coefficients of variation (CVs) of those traits. On the other hand, on the assumption that pollinators are influencing the variation in floral traits (e.g. in pistil length) we expected lower mean phenotypic variation of pollen size and pistil length in pollination-specialist plants than in pollination-generalist ones. A positive correlation between pollen size and pistil length was found for the set of species, but not between the CVs of these traits. This trend was maintained when pollination-specialist plants were analyzed separately, but no statistical significance was obtained for the correlation in pollination-generalist plants. Contrary to our expectations, pollinationspecialist plants did not show less mean intraspecific variation in floral traits than pollination-generalist plants. Therefore, the relationship between pollen size and pistil length among species suggests that the pollination system may be of less importance as a selective force than flower functionality.

Acta Oecologica-international Journal of Ecology, 2010

Reproductive and regenerative traits associated with colonization and persistence ability may det... more Reproductive and regenerative traits associated with colonization and persistence ability may determine plant range size. However, few comparative studies on plant distribution have assessed these traits simultaneously. Pollinator richness and frequency of visits, autonomous self-pollination ability, reproductive output (i.e., reproductive traits), seed bank strategy and seedling density (i.e., regenerative traits) were compared between the narrowly distributed Ipomoea rubriflora O’Donnell (Convolvulaceae)

Object tracking is a long standing problem in vision. While great efforts have been spent to impr... more Object tracking is a long standing problem in vision. While great efforts have been spent to improve tracking performance, a simple yet reliable prior knowledge is left unexploited: the target object in tracking must be an object other than non-object. The recently proposed and popularized objectness measure provides a natural way to model such prior in visual tracking. Thus motivated, in this paper we propose to adapt objectness for visual object tracking. Instead of directly applying an existing objectness measure that is generic and handles various objects and environments, we adapt it to be compatible to the specific tracking sequence and object. More specifically, we use the newly proposed BING [7] objectness as the base, and then train an object-adaptive objectness for each tracking task. The training is implemented by using an adaptive support vector machine that integrates information from the specific tracking target into the BING measure. We emphasize that the benefit of the proposed adaptive objectness, named ADOB-ING, is generic. To show this, we combine ADOBING with seven top performed trackers in recent evaluations. We run the ADOBING-enhanced trackers with their base trackers on two popular benchmarks, the CVPR2013 benchmark (50 sequences) and the Princeton Tracking Benchmark (100 sequences). On both benchmarks, our methods not only consistently improve the base trackers, but also achieve the best known performances. Noting that the way we integrate objectness in visual tracking is generic and straightforward, we expect even more improvement by using tracker-specific objectness.

Acta Oecologica, 2013

This article appeared in a journal published by Elsevier. The attached copy is furnished to the a... more This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier's archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright

Royal Society Open Science, 2019

Women underrepresentation in science has frequently been associated with women being less product... more Women underrepresentation in science has frequently been associated with women being less productive than men (i.e. the gender productivity gap), which may be explained by women having lower success rates, producing science of lower impact and/or suffering gender bias. By performing global meta-analyses, we show that there is a gender productivity gap mostly supported by a larger scientific production ascribed to men. However, women and men show similar success rates when the researchers' work is directly evaluated (i.e. publishing articles). Men's success rate is higher only in productivity proxies involving peer recognition (e.g. evaluation committees, academic positions). Men's articles showed a tendency to have higher global impact but only if studies include self-citations. We detected gender bias against women in research fields where women are underrepresented (i.e. those different from Psychology). Historical numerical unbalance, socio-psychological aspects and c...

Advances in Ecological Research, 2013

PloS one, 2015

Most flowering plants depend on pollinators to reproduce. Thus, evaluating the robustness of plan... more Most flowering plants depend on pollinators to reproduce. Thus, evaluating the robustness of plant-pollinator assemblages to species loss is a major concern. How species interaction patterns are related to species sensitivity to partner loss may influence the robustness of plant-pollinator assemblages. In plants, both reproductive dependence on pollinators (breeding system) and dispersal ability may modulate plant sensitivity to pollinator loss. For instance, species with strong dependence (e.g. dioecious species) and low dispersal (e.g. seeds dispersed by gravity) may be the most sensitive to pollinator loss. We compared the interaction patterns of plants differing in dependence on pollinators and dispersal ability in a meta-dataset comprising 192 plant species from 13 plant-pollinator networks. In addition, network robustness was compared under different scenarios representing sequences of plant extinctions associated with plant sensitivity to pollinator loss. Species with differe...

Organisms Diversity & Evolution, 2009

Twenty-five biotically pollinated plants of the Chaco Serrano Forest (Có rdoba, Argentina) were s... more Twenty-five biotically pollinated plants of the Chaco Serrano Forest (Có rdoba, Argentina) were studied in order to analyze whether 'flower functionality' is related to the relationship between pollen size and pistil length. Because flower functionality may act on the respective mean values of pollen size and pistil length rather than on intraspecific variation in these traits, we expected (1) a high positive correlation between pollen size and pistil length in a set of sympatric species, independent of their degree of pollination specialization or generalization; and (2) no interspecific correlation between the coefficients of variation (CVs) of those traits. On the other hand, on the assumption that pollinators are influencing the variation in floral traits (e.g. in pistil length) we expected lower mean phenotypic variation of pollen size and pistil length in pollination-specialist plants than in pollination-generalist ones. A positive correlation between pollen size and pistil length was found for the set of species, but not between the CVs of these traits. This trend was maintained when pollination-specialist plants were analyzed separately, but no statistical significance was obtained for the correlation in pollination-generalist plants. Contrary to our expectations, pollinationspecialist plants did not show less mean intraspecific variation in floral traits than pollination-generalist plants. Therefore, the relationship between pollen size and pistil length among species suggests that the pollination system may be of less importance as a selective force than flower functionality.

Acta Oecologica-international Journal of Ecology, 2010

Reproductive and regenerative traits associated with colonization and persistence ability may det... more Reproductive and regenerative traits associated with colonization and persistence ability may determine plant range size. However, few comparative studies on plant distribution have assessed these traits simultaneously. Pollinator richness and frequency of visits, autonomous self-pollination ability, reproductive output (i.e., reproductive traits), seed bank strategy and seedling density (i.e., regenerative traits) were compared between the narrowly distributed Ipomoea rubriflora O’Donnell (Convolvulaceae)