This paper presents a method for the estimation of scene structure and camera motion from a sequence of images. This approach is fundamentally new. No computation of optical flow or feature correspondences is required. The method processes image sequences of arbitrary length and exploits the redundancy for a significant reduction in error over time. No assumptions are made about camera motion or surface structure. Both quantities are fully recovered. Our method combines the "direct'' motion vision approach with the theory of recursive estimation. Each step is illustrated and evaluated with results from real images.

This work introduces a direct method called FIXATION for solving the general motion vision problem. This Fixation method results in a constraint equation between translational and rotational velocities that in combination with the Brightness-Change Constraint Equation (BCCE) solves the general motion vision problem, arbitrary motion with respect to an arbitrary rigid environment. Neither Correspondence nor Optical Flow has been used here. Recently Direct Motion Vision methods have used the BCCE for solving the motion vision problem of special motions or environments. In contrast to those solutions, the Fixation method does not put such severe restrictions on the motion or the environment.

Flies use visual cues for a variety of tasks, such as maneuvering through the environment and finding potential mates. Hoverflies, in particular, have very developed eyes and use them to be able to hover mid air and perform fast, elegant movements. The Motion Vision Group, located at the Department of Neuroscience at BMC, Uppsala, studies the motion vision system of the hoverfly brain, using electrophysiology. Experiments are performed by displaying visual stimuli on a screen in front of an immobilized fly, while recording the response from a single neuron with a thin electrode.Until now, the Motion Vision group has been using the open source program VisionEgg to generate the stimuli. VisionEgg is able to display stimuli at high frame rate and has a large set of useful features such as perspective distortion. It also has a lot of drawbacks that makes it desirable to acquire new software. The main drawbacks include it being hard to learn, use and modify, as well as being unable to generate the stimuli needed for some key experiments.This master´s thesis describes the development of software more suited to the lab´s needs. This software should be able to generate some of the stimuli that were impossible to do at the moment, as well as being easy to expand and add upon. The frame rate of the displayed stimuli has to be both high and stable in order to perform high precision experiments.The resulting program is called FlyFly and has been developed iteratively in close cooperation with its end users, ensuring a user friendly end product capable of meeting the lab´s needs. FlyFly is implemented using MatLab and the Psychophysics toolbox with the graphical user interface (GUI) designed with the Guide editor. The GUI is decoupled from the functions drawing the stimuli, making it easy to improve or remove parts altogether. FlyFly is intuitive to use and allows anyone to quickly get started. It allows easy manipulation of series of trials, and supports drawing of multiple objects simultaneously. With the current machine set-up, it displays stimuli at 160 frames per second with few or no dropped frames.FlyFly is currently being used in the lab and will be so for the foreseeable future.

In this paper we show how the theory of dynamical systems can be employed to solve problems in motion vision. In particular we develop algorithms for the recovery of dense depth maps and motion parameters using state space observers or filters. Four different dynamical models of the imaging situation are investigated and corresponding filters/ observers derived. The most powerful of these algorithms recovers depth and motion of general nature using a brightness change constraint assumption. No feature-matching preprocessor is required.

Before corresponding points in images taken with two cameras can be used to recover distances to objects in a scene, one has to determine the position and orientation of one camera relative to the other. This is the classic photogrammetric problem of relative orientation, central to the interpretation of binocular stereo information. Described here is a particularly simple iterative scheme for recovering relative orientation that, unlike existing methods, does not require a good initial guess for the baseline and the rotation.

The goal of this work is to navigate through an office environmentsusing only visual information gathered from four cameras placed onboard a mobile robot. The method is insensitive to physical changes within the room it is inspecting, such as moving objects. Forward and rotational motion vision are used to find doors and rooms, and these can be used to build topological maps. The map is built without the use of odometry or trajectory integration. The long term goal of the project described here is for the robot to build simple maps of its environment and to localize itself within this framework.

Schizophrenia (SCZ) is associated with robust social deficits, which have been shown to precede illness onset and predict functional outcome. As a result, social functioning is an important developmental domain affected by SCZ, which likely has a downstream negative impact on other functional abilities, such as interpersonal relationships and vocational capacity. Patients with SCZ also demonstrate significant visual perceptual deficits; however, a remaining question is whether basic impairment in visual processing gives rise to the deficits observed in social perception. In this context, previous research has shown that biological motion contains relevant social information, such as emotional states and intention, which is easily interpreted by healthy observers. Given that biological motion perception is an important source of social information, and that patients with SCZ have known visual perceptual impairment including motion processing deficits, it is possible that poor biological motion perception meaningfully impacts social perception among individuals with SCZ. While previous studies have documented preliminary evidence of biological motion processing deficits in this population, there is a current lack of understanding regarding the basic visual perceptual mechanisms that may underpin this impairment, including the importance of basic visual motion processing with respect to biological motion. Moreover, the ability of individuals with SCZ to extract relevant social information from biological motion, and its relationship with social perception more generally, have yet to be investigated. Thus, the specific aims guiding the current thesis were to examine whether basic visual motion processes may give rise to biological motion deficits and to examine the ability of individuals with SCZ to extract social information, in the form of emotion, from biological motion. Several experimental tasks were used to examine these aims. Overall, the results from this thesis confirm that individuals with SCZ have difficulty perceiving biological motion; however, this deficit was not specific to biological motion, but instead reflected more widespread visual motion processing deficits, including impairment in global coherent motion perception. Additionally, results from this thesis suggest that individuals with SCZ demonstrated disproportionate difficulty extracting social cues, in the form of emotion, from biological motion, and that this deficit was related to perceiving unambiguous expressions of emotion. In contrast, the discrimination of more subtle or ambiguous emotion was relatively preserved. Moreover, impairment in biological motion processing was found to be unrelated to social perceptual abilities among individuals with SCZ. These experiments provide interesting suggestions regarding clinical approaches to treatment and remediation, although further research is needed to fully understand the brain- behaviour mechanisms underlying SCZ-related deficits in biological motion processing. / Dissertation / Doctor of Philosophy (PhD) / As people navigate though day-to-day life, they encounter many objects in the world that move, such as other people. Research has shown that humans are adept at discriminating human movement and accurate in discerning the emotional states of other people based on this movement. These observations have lead researchers to speculate that, because biological motion is both easy to discriminate and emotionally informative, it plays an essential role in social processing among humans. Research has shown that individuals with Schizophrenia have difficulty understanding social environmental cues, such as the emotions of others. As such, this thesis aims to determine first, whether people with Schizophrenia have difficulty identifying human motion, and second, if they are able to identify emotions embedded within human motion. This thesis will help researchers understand and explain deficits in social perception among people with Schizophrenia.

Flies are highly visually guided animals. In this thesis, I have used hoverflies as a model for studying motion vision. Flies process motion vision in three visual ganglia: the lamina, the medulla, and the lobula complex. In the posterior part of lobula complex, there are around 60 lobula plate tangential cells (LPTCs). Most of LPTCs have large receptive fields where the local direction sensitivity suggests that they function as matched filters to specific types of optic flow. LPTCs connect to descending or neck motor neurons that control wing and head movements, respectively. Therefore, in this thesis I have focused on the electrophysiological responses of LPTCs to gain understanding of visual behaviors in flies. The elementary motion detector (EMD) is a model that can explain the formation of local motion sensitivity. However, responses to higher order motion, where the direction of luminance change is uncorrelated with the direction of movement, cannot be predicted by classic EMDs. Nevertheless, behavior shows that flies can see and track bars with higher order motion cues. I showed (Paper I) that several LPTCs also respond to higher order motion. Many insects, including flies, release octopamine during flight. Therefore, adding octopamine receptor agonists can mimic physical activity. Our study (Paper II) investigated the effect of octopamine on three adaptation components. We found that the contrast gain reduction showed a frequency dependent increase after octopamine stimulation. Since the contrast gain is non-directional, it is likely presynaptic to the LPTC. We therefore believe that octopamine acts on the delay filter in the EMD. In the third paper we describe a novel LPTC. The centrifugal stationary inhibited flicker excited (cSIFE) is excited by flicker and inhibited by stationary patterns. Neither of these responses can be predicted by EMD models. Therefore, we provide a new type of motion detector that can explain cSIFE’s responses (Paper III). During bar tracking, self-generated optic flow may counteract the steering effect by inducing a contradictory optomotor response. Behavior shows that during bar fixation, flies ignore background optic flow. Our study (Paper IV) focus on the different receptive fields of two LPTCs, and relate these to the bar fixation behavior. In the neuron with a small and fronto-dorsal receptive field, we find a higher correlation with bar motion than with background motion. In contrast, the neuron with a larger receptive field shows a higher correlation with background motion.

This thesis describes work from both the University of Cambridge in the lab of Berthold Hedwig and from the HHMI Janelia Research Campus in the lab of Michael Reiser. At the University of Cambridge, my work involved the development and demonstration of a method for electrophoretically delivering dyes and tracers for anatomical and functional imaging into animals that are not amenable to genetic labelling techniques. Using this method in locusts and crickets - model systems of particular interest for their acoustic communication - I successfully delivered polar fluorescent dyes and tracers through the sheath covering the auditory nerve, simultaneously staining both the peripheral sensory structures and the central axonal projections without destroying the nerve's function. I could label neurons which extend far from the tracer delivery site on the nerve as well as local neuron populations through the brain's surface. I used the same method to deliver calcium indicators into central neuropils for in vivo optical imaging of sound-evoked activity, as well as calling song-evoked activity in the brain. The work completed at the Janelia Research Campus began with the development of a modern version of a modular LED display and virtual reality control system to enable research on the visual control of complex behaviors in head-fixed animals. The primary advantages of our newly developed LED-based display over other display technologies are its high-speed operation, brightness uniformity and control, precise synchronization with analog inputs and outputs, and its ability to be configured into a variety of display geometries. Utilizing the system's fast display refresh rates, I conducted the first accurate characterization of the upper limits of the speed sensitivity of Drosophila for apparent motion during flight. I also developed a flexible approach to presenting optic flow scenes for functional imaging of motion-sensitive neurons. Finally, through the on-line analysis of behavioral measures, image rendering, and display streaming with low latency to multi-color (UV/Green) LED panels, I demonstrated the ability to create more naturalistic stimuli and interactive virtual visual landscapes. Lastly, I used this new visual display system to explore a newly discovered cell-type that had been implicated in higher-order motion processing from a large genetic screen of visually-guided behavior deficits. Using genetic silencing and activation methods, and by designing stimuli that modeled the optic flow encountered during different types of self-motion, colleagues in the Reiser lab and I showed that this cell-type - named Lobula Plate Columnar 1 (LPC1) - is required for the stopping behavior of walking flies caused by back-to-front translation motion but is not involved in the rotational optomotor response. Using calcium imaging, I found that LPC1 was selectively excited by back-to-front motion on the eye ipsilateral to the neuron population and inhibited by front-to-back motion on the contralateral eye, demonstrating a simple mechanism for its selectivity to translation over rotation. I also examined an anatomically similar cell type - named Lobula-Lobula Plate Columnar type 1 (LLPC1) - and found that its selectivity results from a similar but opposite calculation for the detection of front-to-back translational motion. The detection of back-to-front motion had previously been hypothesized to be useful for collision avoidance, and this work provides a neural mechanism for how this detection could be accomplished, as well as providing a platform from which to explore the larger network for translation optic flow.