Ivan Ralašić

Mr. Ivan Ralašić

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 July 2018 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
IRIvan Ralašić
Ivan Ralašić
DSDamir Seršić
Damir Seršić
DPDavor Petrinovic
Davor Petrinovic
While numerous works analyze the theoretical background of compressive sensing (CS) and provide rich mathematical theory, few practical implementations exist and they mostly share the same disadvantage of being either too complex or too expensive to implement. As a result of the reported work, a simple measurement setup for CS using off-the-shelf components is proposed. A simple and portable CS architecture consists of a digital camera, a projector, and mechanical integration elements. Extensive analysis of the proposed measurement system and the detailed calibration method is provided. The presented CS method allows for the use of a low-resolution camera with significant lens distortion for data acquisition. The results of our CS method are nondistorted, high-resolution images, which are obtained from the acquired measurement data using sparse optimization.

17 June 2020 in Informatica
Verified
Journal Article
Open access
PDF file
IRIvan Ralašić
Ivan Ralašić
DSDamir Seršić
Damir Seršić
SŠSiniša Šegvić
Siniša Šegvić
Publisher: Vilnius University Institute of Data Science and Digital Technologies, Journal: Informatica, Title: Perceptual Autoencoder for Compressive Sensing Image Reconstruction, Authors: Ivan Ralašić, Damir Seršić, Siniša Šegvić , This paper presents a non-iterative deep learning approach to compressive sensing (CS) image reconstruction using a convolutional autoencoder and a residual learning network. An efficient measurement design is proposed in order to enable training of the compressive sensing models on normalized and mean-centred measurements, along with a practical network initialization method based on principal component analysis (PCA). Finally, perceptual residual learning is proposed in order to obtain semantically informative image reconstructions along with high pixel-wise reconstruction accuracy at low measurement rates.

27 February 2020 in IEEE Access
Verified
Journal Article
Open access
IRIvan Ralašić
Ivan Ralašić
MĐMatea Đonlić
Matea Đonlić
DSDamir Seršić
Damir Seršić
In this paper, a dual imaging technique is used for high-precision reconstruction of an observed 3D scene. In contrast to stereo vision, dual imaging systems use a camera and a projector instead of a camera pair. We propose a multiresolution approach based on the sum-to-one transform, coupled with compressive sensing principles, for efficient estimation of the light transport matrix (LTM). The LTM contains information on both optical systems and the 3D scene. In our setup, the camera sensor is intentionally chosen to be low resolution to prove the future use of inexpensive sensors in nonvisible regions of the light spectrum, as well as the potential design of simplified multiview and light field acquisition systems. We show that a high-precision estimation of the LTM from a reduced set of measurements is possible. Virtual measurements, instead of physical, are conducted to obtain the 3D reconstruction.We show that 3D scene reconstruction from the proposed virtual measurements corresponds with the actual physical acquisition. Moreover, this approach provides much more detail in the reconstruction. The computational complexity of the proposed methods is reduced to such a level that practical implementations are feasible.

21 January 2020 in Automatika
Verified
Journal Article
Open access
NDNicol Dlab
Nicol Dlab
FCFilip Cimermančić
Filip Cimermančić
IRIvan Ralašić
Ivan Ralašić
DSDamir Seršić
Damir Seršić
In contrast to conventional cameras which capture a 2D projection of a 3D scene by integrating the angular domain, light field cameras preserve the angular information of individual light rays by capturing a 4D light field of a scene. On the one hand, light field photography enables powerful post-capture capabilities such as refocusing, virtual aperture, depth sensing and perspective shift. On the other hand, it has several drawbacks, namely, high-dimensionality of the captured light fields and a fundamental trade-off between spatial and angular resolution in the camera design. In this paper, we propose a compressive sensing approach to light field acquisition from a sub-Nyquist number of samples. Using an off-the-shelf measurement setup consisting of a digital projector and a Lytro Illum light field camera, we demonstrate the efficiency of the compressive sensing approach by improving the spatial resolution of the acquired light field. This paper presents a proof of concept with a simplified 3D scene as the scene of interest. Results obtained by the proposed method show significant improvement in the spatial resolution of the light field as well as preserved post-capture capabilities.