In many applications, such as conversational agents, virtual reality, movies, and games, animated facial expressions of computer-generated (CG) characters are used to communicate, teach, or entertain. With an increased demand for CG characters, it is important to animate accurate, realistic facial expressions because human facial expressions communicate a wealth of information. However, realistically animating faces is challenging and time-consuming for two reasons. First, human observers are adept at detecting anomalies in realistic CG facial animations. Second, traditional animation techniques based on keyframing sometimes approximate the dynamics of facial expressions or require extensive artistic input while high-resolution performance capture techniques are cost prohibitive. In this thesis, we develop a framework to explore representations of two key facial expressions, blinks and smiles, and we show that data-driven models are needed to realistically animate these expressions. Our approach relies on utilizing high-resolution performance capture data to build models that can be used in traditional keyframing systems. First, we record large collections of high-resolution dynamic expressions through video and motion capture technology. Next, we build expression-specific models of the dynamic data properties of blinks and smiles. We explore variants of the model and assess whether viewers perceive the models as more natural than the simplified models present in the literature. In the first part of the thesis, we build a generative model of the characteristic dynamics of blinks: fast closing of the eyelids followed by a slow opening. Blinks have a characteristic profile with relatively little variation across instances or people. Our results demonstrate the need for an accurate model of eye blink dynamics rather than simple approximations, as viewers perceive the difference. In the second part of the thesis, we investigate how spatial and temporal linearities impact smile genuineness and build a model for genuine smiles. Our perceptual results indicate that a smile model needs to preserve temporal information. With this model, we synthesize perceptually genuine smiles that outperform traditional animation methods accompanied by plausible head motions. In the last part of the thesis, we investigate how blinks synchronize with the start and end of spontaneous smiles. Our analysis shows that eye blinks correlate with the end of the smile and occur before the lip corners stop moving downwards. We argue that the timing of blinks relative to smiles is useful in creating compelling facial expressions. Our work is directly applicable to current methods in animation. For example, we illustrate how our models can be used in the popular framework of blendshape animation to increase realism while keeping the system complexity low. Furthermore, our perceptual results can inform the design of realistic animation systems by highlighting common assumptions that over-simplify the dynamics of expressions.

<p>Electrooculogram (EOG) data was used to develop, adjust and validate a method for drowsiness detection in drivers. The drowsiness detection was based on changes in blink behaviour and classification was made on a four graded scale. The purpose was to detect early signs of drowsiness in order to warn a driver. MATLAB was used for implementation. For adjustment and validatation, two different reference measures were used; driver reported ratings of drowsiness and an electroencephalogram (EEG) based scoring scale. A correspondence of 70 % was obtained between the program and the self ratings and 56 % between the program and the EEG based scoring scale. The results show a possibility to detect drowsiness by analyzing blink behaviour changes, but that inter-individual differences need to be considered. It is also difficult to find a comparable reference measure. The comparability of the blink based scale and the EEG based scale needs further investigation.</p>

Electrooculogram (EOG) data was used to develop, adjust and validate a method for drowsiness detection in drivers. The drowsiness detection was based on changes in blink behaviour and classification was made on a four graded scale. The purpose was to detect early signs of drowsiness in order to warn a driver. MATLAB was used for implementation. For adjustment and validatation, two different reference measures were used; driver reported ratings of drowsiness and an electroencephalogram (EEG) based scoring scale. A correspondence of 70 % was obtained between the program and the self ratings and 56 % between the program and the EEG based scoring scale. The results show a possibility to detect drowsiness by analyzing blink behaviour changes, but that inter-individual differences need to be considered. It is also difficult to find a comparable reference measure. The comparability of the blink based scale and the EEG based scale needs further investigation.

Visual field tests are routinely used for the detection and management of glaucoma. This thesis focuses on monitoring the vigilance during perimetry with pupil dynamics (pupillary movements, blinks) in glaucoma patients and the performance of using subsets of standard test patterns at detecting glaucomatous field loss. It describes the spatial pattern of visual field changes in diabetic retinopathy patients after treatment and the usage of multispectral imaging of the optic nerve head in glaucoma patients.The first investigation evaluated the use of pupillary parameters (pupil size and amplitude of pupillary oscillations) to monitor vigilance during perimetry in comparison to electroencephalography. A significant correlation was observed between the decreased pupil size and increased slow brain activities (theta and alpha). The inter-test threshold variability correlated with the pupil miosis.Another investigation explored blinks during a perimetric test and their relationship to the variability of threshold sensitivity in glaucoma patients. A wide range of blink frequencies was observed in glaucoma patients during perimetry and the timing of blinks was correlated with the stimuli presentations when they were supra-threshold. No significant relationship was observed between the threshold sensitivity variability and blink parameters (frequency, duration and microsleep numbers).The third investigation evaluated the use of subset of the 24-2 visual field test pattern at detecting glaucomatous field loss. A high sensitivity and specificity was observed when using subsets of the 24-2 pattern and test locations optimized by the positive predictive values showed a better performance than the test locations in random sequences.With the usage of a self-organized mapping method, nine spatial patterns of visual field loss in untreated diabetic retinopathy patients were classified. After laser treatment, a significant improvement on the spatial patterns of field loss and global indices was found.In the multispectral imaging study, the haemoglobin dependent differential light absorption (DLA) properties of the glaucomatous optic nerve head was evaluated with multispectral imaging. Change of DLA after medical hypotensive treatment was graded by five experts and no difference was observed in most of the patients.

La validité des simulateurs de conduite est une question essentielle pour étayer et valoriser les travaux visant à mieux comprendre les comportements de conduite. Alors que les dimensions subjective et comportementale de la validité ont été régulièrement étudiées, elles font face à certaines limites. A l’opposé, rares sont les études qui se sont intéressées à la validité psychologique des simulateurs. Cette dimension compare l'implication des mécanismes qui sous-tendent les comportements entre conduite réelle et virtuelle. De plus, très peu d’études ont confronté les différentes dimensions de la validité, alors que cela pourrait contribuer au développement d’une métrologie des environnements virtuels. C’est précisément à ce niveau que se situe ce travail de thèse, au cours duquel des mesures classiques de la validité comportementale (vitesse, contrôle latéral) ont été confrontées à des mesures de la validité psychologique examinée sous l’angle de la charge mentale en prenant notamment comme indicateur les clignements oculaires. L’objectif principal était de déterminer si le niveau de charge mentale diffère entre conduite sur route et conduite sur simulateur dans des cas où les comportements observés ne permettent pas de faire de distinction.Pour répondre à cette question, ce travail de thèse a été organisé autour de trois expériences réalisées sur simulateur de conduite et d’une étude sur route réelle. Les deux premières expériences visaient à mieux cerner l’effet de certains facteurs inhérents à la conduite sur route ouverte (trafic, environnement) sur les comportements de conduite et la charge mentale. La troisième expérience était quant à elle destinée à comparer la charge mentale induite par la conduite réelle et la conduite sur simulateur « bas coût », en confrontant cette mesure de la validité psychologique aux mesures comportementales classiques. Enfin, la quatrième expérience a porté sur les effets du retour d’effort du volant sur cette même charge mentale.Les principaux résultats ont mis en évidence un niveau de charge supérieur en conduite simulée par comparaison à la conduite en situation réelle, alors que certaines mesures comportementales (vitesse) n’étaient pas différentes. Cette charge accrue sur simulateur ne semble cependant pas trouver son origine dans le retour d’effort au volant, les modalités de retour testées n’ayant pas eu d’effet sur les indicateurs de la charge. Au final, ce travail confirme que la prise en compte de la validité psychologique, examinée ici sous l’angle de la charge mentale, présente un intérêt dans une démarche d’évaluation, dans le but de mieux cerner le positionnement des utilisateurs face à un dispositif de réalité virtuelle. Il ouvre ainsi des perspectives pour améliorer la validité des simulateurs de conduite. / The validity of driving simulators is an essential subject to support and highlight the works aiming to understand driving behaviours more thoroughly. While the subjective and behavioural dimensions of that validity have often been studied, they encounter a few limits. On the other hand, studies about simulators’ validity are fairly rare. This dimension compares the implication of mechanisms inherent in behaviours between real and virtual driving. Furthermore, very few studies have considered the various dimensions of validity at once, whereas it could contribute to the development of a metrology for virtual environments. This thesis is precisely about this, with confrontations between classical measurements of behavioural validity (speed, lateral control) and measurements of psychological validity, examined from the viewpoint of mental workload - using indicators such as eye blinking. The main objective was to ascertain whether the mental workload levels vary between road driving and simulated driving when the observed behaviours aren’t relevant to make a clear difference.To answer this question, this thesis has been organised around three experiences carried out on driving simulators and a comparison between actual road driving and low-cost simulated driving. Two of those experiences were conducted to finely analyse the effect of some factors inherent to open road driving (traffic, environment) on driving behaviours and mental workload. The third experience was made to compare the mental workload induced by real driving and low-cost simulator driving, by putting this psychological validity measurement against classical behavioural measurements. Lastly, the fourth experience focused on the effects of the driving wheel’s force-feedback on this same mental workload.The main results revealed a higher workload level on simulated driving than on real driving, while some behavioural measurements (speed) were not different. This increased workload with the simulator does not seem to stem from the wheel’s force-feedback, since the tested feedback methods did not have a repercussion on the workload levels. In the end, this work confirms that taking psychological validity - examined here from a mental workload viewpoint - does have an interest within an evaluation process to analyse in a finer fashion the mental state of users when faced with a virtual reality system. It creates prospects to improve the validity of driving simulators.

A relação trabalho e visão é tão remota que determinou a adaptação da espécie humana ao meio em que está inserida. Demonstrou-se que trabalhar em um ambiente confortável, sem doenças ocular, favorece o desempenho visual e aumenta a produtividade. Conhecer os efeitos do monitor de computador (VDT) sobre o sistema visual significa planejar melhor o ambiente de trabalho e a atividade laboral para se obter máximo conforto e produtividade. Este estudo teve como objetivo geral analisar as alterações do piscar, do filme lacrimal e da superfície ocular induzidas pelo uso de monitor de computador (VDT). Para sua realização efetivou-se um estudo transversal, descritivo, observacional, analítico, realizado na Volkswagen do Brasil, São José dos Pinhais (PR), tendo como amostra, a população de funcionários do setor administrativo que utiliza VDT, com idade acima de 18 anos, sem doenças da superfície ocular, sistema palpebral e lacrimal. O ambiente de trabalho foi caracterizado pela temperatura, grau de umidade, velocidade do ar e iluminamento. Preencheram os critérios de inclusão 108 funcionários (idade média 34,1±7,88 anos e tempo mínimo de utilização do VDT 6 horas/dia). Foram avaliados as sintomatologias, filmagem da face para mensuração do tempo entre piscadas na situação de conversação, exame do olho externo e biomicroscopia das pálpebras, conjuntiva e córnea, filmagem videoceratoscópica para determinar tempo de ruptura do filme lacrimal, avaliação da quantidade de lágrima do lago lacrimal pelo Zone-Quick Phenol Red Thread tear test, avaliação do epitélio corneano pelo teste da lissamina verde, filmagem da face dos sujeitos para mensuração do tempo entre piscadas em leitura de VDT. Ao final da jornada de trabalho, foram feitas a reavaliação do tempo de ruptura do filme lacrimal da quantidade de lágrima e a avaliação do epitélio corneano. Compararamse o tempo entre as piscadas em conversação e em uso de VDT e o tempo de ruptura do filme lacrimal, antes e ao final da jornada de trabalho. A comparação desses tempos classificou os sujeitos como expostos, com tempo médio de ruptura do filme lacrimal menor que o tempo médio entre piscadas e não expostos com tempo médio de ruptura do filme lacrimal maior que o tempo médio entre piscadas. Nas condições do estudo, em jornada de trabalho com tempo mínimo de 6 horas de uso de VDT, houve aumento da exposição da superfície ocular causada pelo aumento do tempo entre piscadas. O aumento da exposição da superfície ocular foi causado pela diminuição do tempo de ruptura do filme lacrimal constatado ao final da jornada de trabalho. O tempo entre piscadas na situação de leitura em VDT que foi significantemente maior que na situação de conversação. O tempo de ruptura do filme lacrimal, ao final da jornada de trabalho, foi significativamente menor que em seu início. A exposição ocular observada durante a jornada de trabalho com usuários de VDT, nas condições deste estudo, não causou alteração epitelial detectável ao teste da lissamina verde. Não houve alteração do volume lacrimal em olhos com exposição, antes da jornada de trabalho com VDT, nem houve alteração do volume lacrimal em olhos com exposição ocular, após a jornada de trabalho com VDT. / The relationship between work and vision is so remote that has determined the adaptation of human beings to the environment in which they are inserted. It has been demonstrated that working in a pleasant environment, without any ocular disease, has favored visual performance and increased productivity. Knowing the effects of video terminal display (VDT) on the visual system helps plan labor environments and labor activities more effectively so that optimal comfort and productivity are obtained. This transversal study, performed at Volkswagen do Brasil, São José dos Pinhais, PR, has the objective to investigate the influence of VDT labor activities on blinking, tear film and ocular surface, ocular exposure and amount of tear in the lacrimal lake. The population sample comprised VDT administrative department employees older than 18, without diseases on the ocular surface, tear film and blinking system. The evaluation of the working environment included measurements of temperature, humidity, air speed and lighting. One hundred and eight employees met the study criteria (mean age 34.1±7.88 and minimal VDT use time of 6 hours/day). Several evaluations were performed: the ocular surface symptomatology; the measurement of the blink time in conversation situations; the examination of the ocular surface and lacrimal film; a videokeratoscopy to establish the tear film break-up time; evaluation of the amount of tears of the lacrimal lake by means of the Zone-Quick Phenol Red Thread tear test; evaluation of the corneal epithelium by means of the lissamine green test; and determination of the blink time in VDT reading situations. At the end of the working period, the amount of tears and the corneal epithelium structure were re-evaluated. Comparisons between the blink time in conversation and VDT reading situations were performed before and at the end of the working day. The comparison data obtained classifies the subjects into two categories: exposed, with a tear film break-up time shorter than the mean time between blinks; and non-exposed, with tear film break-up time longer than the mean time between blinks. The findings obtained reveal that there was an increase in the ocular surface exposure caused by the increase in the time between blinks during the working period and increase of the ocular surface exposure caused by the decrease in the tear film break-up time determined at the end of the working day. The time between blinks at VDT reading situations was significantly longer than at conversation situations. The tear film break-up time at the end of the working day was significantly shorter than at the beginning. The ocular exposure during the VDT use has not caused epithelium alterations detectable at the lissamine green test. There were no alterations in the amount of tears in the eyes which underwent ocular exposure after VDT use situations.

A relação trabalho e visão é tão remota que determinou a adaptação da espécie humana ao meio em que está inserida. Demonstrou-se que trabalhar em um ambiente confortável, sem doenças ocular, favorece o desempenho visual e aumenta a produtividade. Conhecer os efeitos do monitor de computador (VDT) sobre o sistema visual significa planejar melhor o ambiente de trabalho e a atividade laboral para se obter máximo conforto e produtividade. Este estudo teve como objetivo geral analisar as alterações do piscar, do filme lacrimal e da superfície ocular induzidas pelo uso de monitor de computador (VDT). Para sua realização efetivou-se um estudo transversal, descritivo, observacional, analítico, realizado na Volkswagen do Brasil, São José dos Pinhais (PR), tendo como amostra, a população de funcionários do setor administrativo que utiliza VDT, com idade acima de 18 anos, sem doenças da superfície ocular, sistema palpebral e lacrimal. O ambiente de trabalho foi caracterizado pela temperatura, grau de umidade, velocidade do ar e iluminamento. Preencheram os critérios de inclusão 108 funcionários (idade média 34,1±7,88 anos e tempo mínimo de utilização do VDT 6 horas/dia). Foram avaliados as sintomatologias, filmagem da face para mensuração do tempo entre piscadas na situação de conversação, exame do olho externo e biomicroscopia das pálpebras, conjuntiva e córnea, filmagem videoceratoscópica para determinar tempo de ruptura do filme lacrimal, avaliação da quantidade de lágrima do lago lacrimal pelo Zone-Quick Phenol Red Thread tear test, avaliação do epitélio corneano pelo teste da lissamina verde, filmagem da face dos sujeitos para mensuração do tempo entre piscadas em leitura de VDT. Ao final da jornada de trabalho, foram feitas a reavaliação do tempo de ruptura do filme lacrimal da quantidade de lágrima e a avaliação do epitélio corneano. Compararamse o tempo entre as piscadas em conversação e em uso de VDT e o tempo de ruptura do filme lacrimal, antes e ao final da jornada de trabalho. A comparação desses tempos classificou os sujeitos como expostos, com tempo médio de ruptura do filme lacrimal menor que o tempo médio entre piscadas e não expostos com tempo médio de ruptura do filme lacrimal maior que o tempo médio entre piscadas. Nas condições do estudo, em jornada de trabalho com tempo mínimo de 6 horas de uso de VDT, houve aumento da exposição da superfície ocular causada pelo aumento do tempo entre piscadas. O aumento da exposição da superfície ocular foi causado pela diminuição do tempo de ruptura do filme lacrimal constatado ao final da jornada de trabalho. O tempo entre piscadas na situação de leitura em VDT que foi significantemente maior que na situação de conversação. O tempo de ruptura do filme lacrimal, ao final da jornada de trabalho, foi significativamente menor que em seu início. A exposição ocular observada durante a jornada de trabalho com usuários de VDT, nas condições deste estudo, não causou alteração epitelial detectável ao teste da lissamina verde. Não houve alteração do volume lacrimal em olhos com exposição, antes da jornada de trabalho com VDT, nem houve alteração do volume lacrimal em olhos com exposição ocular, após a jornada de trabalho com VDT. / The relationship between work and vision is so remote that has determined the adaptation of human beings to the environment in which they are inserted. It has been demonstrated that working in a pleasant environment, without any ocular disease, has favored visual performance and increased productivity. Knowing the effects of video terminal display (VDT) on the visual system helps plan labor environments and labor activities more effectively so that optimal comfort and productivity are obtained. This transversal study, performed at Volkswagen do Brasil, São José dos Pinhais, PR, has the objective to investigate the influence of VDT labor activities on blinking, tear film and ocular surface, ocular exposure and amount of tear in the lacrimal lake. The population sample comprised VDT administrative department employees older than 18, without diseases on the ocular surface, tear film and blinking system. The evaluation of the working environment included measurements of temperature, humidity, air speed and lighting. One hundred and eight employees met the study criteria (mean age 34.1±7.88 and minimal VDT use time of 6 hours/day). Several evaluations were performed: the ocular surface symptomatology; the measurement of the blink time in conversation situations; the examination of the ocular surface and lacrimal film; a videokeratoscopy to establish the tear film break-up time; evaluation of the amount of tears of the lacrimal lake by means of the Zone-Quick Phenol Red Thread tear test; evaluation of the corneal epithelium by means of the lissamine green test; and determination of the blink time in VDT reading situations. At the end of the working period, the amount of tears and the corneal epithelium structure were re-evaluated. Comparisons between the blink time in conversation and VDT reading situations were performed before and at the end of the working day. The comparison data obtained classifies the subjects into two categories: exposed, with a tear film break-up time shorter than the mean time between blinks; and non-exposed, with tear film break-up time longer than the mean time between blinks. The findings obtained reveal that there was an increase in the ocular surface exposure caused by the increase in the time between blinks during the working period and increase of the ocular surface exposure caused by the decrease in the tear film break-up time determined at the end of the working day. The time between blinks at VDT reading situations was significantly longer than at conversation situations. The tear film break-up time at the end of the working day was significantly shorter than at the beginning. The ocular exposure during the VDT use has not caused epithelium alterations detectable at the lissamine green test. There were no alterations in the amount of tears in the eyes which underwent ocular exposure after VDT use situations.

L'entrée visuelle rétinienne est discontinue. D'une part les saccades causent un énorme mouvement de l'image sur la rétine 3 à 4 fois par seconde, qui devrait résulter en un floutage des hautes fréquences spatiales et une forte impression de mouvement. D'autre part, les clignements des yeux induisent une diminution temporaire drastique de la luminance toutes les 3 à 4 secondes. Dans des conditions de vision écologiques, ces conséquences visuelles des saccades et des clignements des yeux ne sont pas consciemment perçues et le monde extérieur semble continu et net : deux phénomènes que l'on peut désigner sous le terme d'omission saccadique et d'omission des clignements des yeux. Dans cette thèse, nous avons voulu mieux comprendre comment le système visuel s'accommode de ces interruptions et quels sont les mécanismes qui contribuent à la continuité perceptive autour des saccades et des clignements des yeux. Deux principaux éléments pourraient contribuer à ces omissions : l'entrée visuelle elle-même et un mécanisme extra-rétinien qui informerait le cerveau de l'interruption à venir qui agirait en modifiant le traitement de l'information autour des saccades et des clignements des yeux. Dans une première série d'expériences, nous avons étudié les caractéristiques du masquage du smear saccadique, c'est à dire dans quelle mesure la présence d'images pré et post saccadiques nettes permet de rendre compte de l'omission du smear saccadique. Plus précisément, nous avons élaboré une méthode de mesure objective du masquage du smear et examiné son étendue spatiale et son origine périphérique ou centrale. A l'aide de cette nouvelle méthode, nous avons répliqué les résultats de masquage du smear et mis en évidence que ce masquage a lieu après le site d'interaction binoculaire et résiste à des séparations spatiales entre smear et masque jusqu'à 6 deg. Dans une deuxième étude nous avons comparé la sensibilité à des réseaux sinusoïdaux de basse fréquence spatiale autour des saccades et en fixation lorsque l'entrée visuelle simule les conséquences visuelles des saccades. De plus, nous avons cherché à établir si la plus importante diminution de la sensibilité observée pour de vraies saccades en comparaison des saccades simulées peut être expliquée par les propriétés cinématiques des mouvements oculaires. L'objectif de la troisième étude était de déterminer si le masquage est suffisant pour rendre compte de l'absence de percept de mouvement autour des saccades. Pour ce faire, nous avons présenté en fixation, un stimulus dont le contenu fréquentiel est similaire à celui des scènes naturelles. Ce stimulus était présenté en mouvement avec un profil similaire à celui d'une saccade. Il pouvait être précédé et suivi de l'image statique avant et après le mouvement. Les résultats indiquent que l'amplitude du mouvement perçu diminue considérablement en présence des masques, sans toutefois annuler totalement tout percept de mouvement pour des longues durées de masques. Dans une dernière série d'études nous nous sommes intéressés à la perception de la durée autour des clignements des yeux. Dans la première expérience nous avons quantifié la contribution de la durée d'un clignement des yeux à la durée d'une période d'obscurité plus longue et dans la seconde expérience, nous avons étudié la perception de la durée d'un objet interrompu ou non par un clignement des yeux. Les résultats de ces deux expériences suggèrent l'implication d'un mécanisme extra-rétinien qui supprime la durée perçue de la période d'obscurité causée par les clignements des yeux mais pas la durée des objets visuels chevauchés par le clignement. Pris dans leur ensemble ces résultats précisent notre compréhension des contributions relatives des mécanismes rétiniens et extra-rétiniens à l'omission saccadique et l'omission des clignements des yeux. / The retinal input is discontinuous. On the one hand saccades, that occur 3-4 times a minute, cause a huge motion of the image on the retina that should result in smearing of the high frequencies of the image and perceived motion. On the other hand eye blinks induce drastic transient decreases in luminance every 3-4 seconds. Under real-world conditions, those visual consequences of saccades and blinks are barely noticed and the world appears continuous and sharp: two phenomena that can be referred to as saccadic and blink omission. In this thesis we were interested in understanding how the visual system deals with these interruptions and which mechanisms contribute to perceived continuity around saccades and blinks. Two main elements could contribute to those omissions: the visual input itself and an extra-retinal mechanism informing the brain of the impending interruption that would affect information processing around saccades and blinks. In a first series of experiments we studied the characteristics of masking of the saccadic smear, the extent to which clear and still pre- and post-saccadic images are responsible for the perceptual omission of saccadic smear. In particular, we designed an objective method to measure smear masking and studied its spatial extent and whether it is of peripheral or central origin. We replicated previous results of saccadic masking with this new method and found that smear masking seems to take place after the site of binocular interaction and survives separations between smear and mask as much as 6 deg. In a second study we compared sensitivity to low-frequency gratings around saccades and in fixation when the visual input simulates the visual consequences of saccades. Moreover we tried to determine whether the greater decrease in sensitivity around real, as compared to simulated, saccades that we found could be accounted for by the cinematic properties of the eye movement. The goal of the third study was to determine if masking was sufficient to explain the lack of perceived motion during saccades. To do that we presented, during fixation, a natural scene-like stimulus moving at saccadic speeds that could be preceded and followed by the initial or final static image. Results indicate that the amplitude of perceived motion considerably decreased in the presence of pre- and post-masks, even though motion was still perceived for long mask durations. In a final series of studies, we probed duration perception around blinks. In a first experiment we quantified the contribution of the duration of a blink to a longer period of darkness and in a second experiment we tested the perceived duration of an object interrupted or not by a blink. Results suggest the involvement of an extra-retinal mechanism that suppresses the perceived duration of the darkness caused by the blink, but not the duration of visual objects that straddle the blink. Taken together these results refine our understanding of the relative contributions of retinal and extra-retinal mechanisms to saccadic and blink omission.