Flying at speed through complex environments is a challenging task that has been performed successfully by insects since the Carboniferous, but which remains a challenge for robotic and autonomous systems. Insects navigate the world using optical flow sensed by their compound eyes, which they process using a deep neural network weighing just a few milligrams. Deploying an insect-inspired network architecture in computer vision could therefore enable more efficient and effective ways of estimating structure and self-motion using optical flow. Training a bio-informed deep network to implement these tasks requires biologically relevant training, test, and validation data. To this end, we introduce FlyView, a novel bio-informed truth dataset for visual navigation. This simulated dataset is rendered using open source 3D scenes in which the observer's position is known at every frame, and is accompanied by truth data on depth, self-motion, and motion flow. This dataset comprising 42,475 frames has several key features that are missing from existing optical flow datasets, including: (i) panoramic cameras with a monocular and binocular field of view matched to that of a fly's compound eyes; (ii) dynamically meaningful self-motion modelled on motion primitives, or the 3D trajectories of drones and flies; and (iii) complex natural and indoor environments including reflective surfaces.

Figure 1: (a) Field of view of the blowfly Calliphora vicina mapped onto a panoramic view of an indoor scene. Blue and red borders outline the visual field of the left and right eyes; tinted area denotes area of binocular overlap.

Flying at speed through complex environments is a challenging task that has been performed successfully by insects since the Carboniferous, but which remains a challenge for robotic and autonomous systems. Insects navigate the world using optical flow sensed by their compound eyes, which they process using a deep neural network weighing just a few milligrams. Deploying an insect-inspired network architecture in computer vision could therefore enable more efficient and effective ways of estimating structure and self-motion using optical flow. Training a bio-informed deep network to implement these tasks requires biologically relevant training, test, and validation data. To this end, we introduce FlyView, a novel bio-informed truth dataset for visual navigation.

Cameras inspired by the compound eyes of insects enable an extremely wide field of view without expensive lenses, potentially offering cheap, simple and lightweight visual sensors for navigation or tracking in robots and driverless cars. Insects like dragonflies have eyes that, in pairs, provide an almost 360-degree field of vision and help them to deftly evade predators. Their eyes are composed of many ommatidia, which are essentially tubes with a simple lens at one end and a basic photoreceptor at the other. Their vision is made up of pixel-like inputs from large bundles of these ommatidia. Creating cameras that can affordably achieve the same thing, either by covering a hemisphere with image sensors or by creating multiple lenses to direct light onto a central sensor, has proved challenging.