High Frame Rate Egomotion Estimation

Abstract

Abstract. In this paper, we present an algorithm for doing high frame rate egomotion estimation. We achieve this by using a basis flow model, along with a novel inference algorithm, that uses spatio-temporal gradients, foregoing the computation of the slow and noisy optical flow. The inherent linearity in our model allows us to achieve fine grained parallelism. We demonstrate this by running our algorithm on GPUs to achieve egomotion estimation at 120Hz.Image motion is tightly coupled with the camera egomotion and depth of the scene. Hence, we validate our approach by using the egomotion estimate to compute the depth of a static scene. Our applications are aimed towards autonomous navigation scenarios where, it is required to have a quick estimate of the state of the vehicle, while freeing up computation time for higher level vision tasks.

Fast Optical Flow Using Dense Inverse Search

“…Recently, only very few works aimed at balancing accuracy and run-time in favor of efficiency [17,48,54], or employed massively parallelized dedicated hardware to achieve acceptable runtimes [4,40,18]. In contrast to this, recently it has been noted for several computer vision tasks [24,44,8,6], that it is often desirable to trade-off powerful but complex algorithms for simple and efficients methods, and rely on high frame-rates and smaller search spaces for good accuracy.…”

Primate-inspired vehicle navigation using optic flow and mental rotations

“…This is advantageous because it makes the estimate of motion more accurate and also scalable to GPU architectures that enable us to do the motion estimation at high frame rates (120Hz) in real time. However, the details of the motion estimation itself are beyond the scope of this paper, but can be found here 22 . In this paper, we are interested in using this estimate to obtain the translational flow for performing the task of mental rotations.…”