Alireza Ghasemi

Scale-invariant representation of light field images for object recognition and tracking

We propose a scale-invariant feature descriptor for representation of light-field images. The proposed descriptor can significantly improve tasks such as object recognition and tracking on images taken with recently popularized light field cameras. We test our proposed representation using various light field images of different types, both synthetic and real. Our experiments showvery promising results in terms of retaining invariance under various scaling transformations.

Detecting planar surface using a light-field camera with application to distinguishing real scenes from printed photos

We propose a novel approach for detecting printed photos from natural scenes using a light-field camera. Our approach exploits the extra information captured by a light-field camera and the multiple views of scene in order to infer a compact feature vector from the variance in the distribution of the depth of the scene. We then use this feature for robust detection of printed photos. Our algorithm can be used in person-based authentication applications to avoid intruding the system using a facial photo. Our experiments show that the energy of the gradients of points in the epipolar domain is highly discriminative and can be used to distinguish printed photos from original scenes.

On the Accuracy of Point Localisation in a Circular Camera-Array

Although many advances have been made in light-field and camera-array image processing, there is still a lack of thorough analysis of the localisation accuracy of different multi-camera systems. By considering the problem from a frame-quantisation perspective, we are able to quantify the point localisation error of circular camera configurations. Specifically, we obtain closed form expressions bounding the localisation error in terms of the parameters describing the acquisition setup. These theoretical results are independent of the localisation algorithm and thus provide fundamental limits on performance. Furthermore, the new frame-quantisation perspective is general enough to be extended to more complex camera configurations.

Hyperspectral imaging using a commercial light-field camera (Conference Presentation)

Hyperspectral imaging allows the collection of both spectral and spatial information. This modality is naturally fitted for object and material identification or detection processes, and has encountered a large success in the agriculture and food industries to name a few. In snapshot spectral imaging, the 3D cube of images is taken in one shot, with the advantage that dynamic scenes can be analyzed. The simplest way to make a hyperspectral camera is to put an array of wavelength filters on the detector and then integrate this detector with standard camera objectives. The technical challenge is to make arrays of N wavelength filters and repeat this sequence up to 100‘000 times across the detector array, where each individual filter is matched to the pixel size and can be as small as a few microns. In this work, we generate the same effect with just one N wavelength filter array which is then multiplied and imaged optically onto the detector to achieve the same effective filter array. This was first outlined by Levoy and Hoystmeyer using microlens arrays in a light field camera (plenoptics 1.0). Instead of building our own light field camera we used an existing commercial camera, Lytro™ and used it as the engine for our telecentric hyperspectral camera. In addition, the tools to extract and rebuild the raw data from the Lytro™ camera were developed. We demonstrate reconstructed hyperspectral images with 9 spectral channels and show how this can be increased to achieve 81 spectral channels in a single snapshot.

Shape: Linear-time camera pose estimation with quadratic error-decay

We propose a novel camera pose estimation or perspective-n-point (PnP) algorithm, based on the idea of consistency regions and half-space intersections. Our algorithm has linear time-complexity and a squared reconstruction error that decreases at least quadratically, as the number of feature point correspondences increase., Inspired by ideas from triangulation and frame quantisation theory, we define consistent reconstruction and then present SHAPE, our proposed consistent pose estimation algorithm. We compare this algorithm with state-of-the-art pose estimation techniques in terms of accuracy and error decay rate. The experimental results verify our hypothesis on the optimal worst-case quadratic decay and demonstrate its promising performance compared to other approaches.