arXiv subjects

Muhammad Waleed

Publications and source records attributed to Muhammad Waleed.

4 recordsLinked to original sources

arXiv2024

Camera Calibration through Geometric Constraints from Rotation and Projection Matrices

The process of camera calibration involves estimating the intrinsic and extrinsic parameters, which are essential for accurately performing tasks such as 3D reconstruction, object tracking and augmented reality. In this work, we propose a novel constraints-based loss for measuring the intrinsic (focal length: $(f_x, f_y)$ and principal point: $(p_x, p_y)$) and extrinsic (baseline: ($b$), disparity: ($d$), translation: $(t_x, t_y, t_z)$, and rotation specifically pitch: $(\theta_p)$) camera parameters. Our novel constraints are based on geometric properties inherent in the camera model, including the anatomy of the projection matrix (vanishing points, image of world origin, axis planes) and the orthonormality of the rotation matrix. Thus we proposed a novel Unsupervised Geometric Constraint Loss (UGCL) via a multitask learning framework. Our methodology is a hybrid approach that employs the learning power of a neural network to estimate the desired parameters along with the underlying mathematical properties inherent in the camera projection matrix. This distinctive approach not only enhances the interpretability of the model but also facilitates a more informed learning process. Additionally, we introduce a new CVGL Camera Calibration dataset, featuring over 900 configurations of camera parameters, incorporating 63,600 image pairs that closely mirror real-world conditions. By training and testing on both synthetic and real-world datasets, our proposed approach demonstrates improvements across all parameters when compared to the state-of-the-art (SOTA) benchmarks. The code and the updated dataset can be found here: https://github.com/CVLABLUMS/CVGL-Camera-Calibration

cs.CV↗

arXiv2021

Enhanced optical trapping via structured scattering

Interferometry can completely redirect light, providing the potential for strong and controllable optical forces. However, small particles do not naturally act like interferometric beamsplitters, and the optical scattering from them is not generally thought to allow efficient interference. Instead, optical trapping is typically achieved via deflection of the incident field. Here we show that a suitably structured incident field can achieve beamsplitter-like interactions with scattering particles. The resulting trap offers order-of-magnitude higher stiffness than the usual Gaussian trap in one axis, even when constrained to phase-only structuring. We demonstrate trapping of 3.5 to 10.0~$\mu$m silica spheres, achieving stiffness up to 27.5$\pm$4.1 times higher than is possible using Gaussian traps, and two orders of magnitude higher measurement signal-to-noise ratio. These results are highly relevant to many applications, including cellular manipulation, fluid dynamics, micro-robotics, and tests of fundamental physics.

physics.optics↗

arXiv2020

Ultrafast viscosity measurement with ballistic optical tweezers

Viscosity is an important property of out-of-equilibrium systems such as active biological materials and driven non-Newtonian fluids, and for fields ranging from biomaterials to geology, energy technologies and medicine. However, noninvasive viscosity measurements typically require integration times of seconds. Here we demonstrate a four orders-of-magnitude improvement in speed, down to twenty microseconds, with uncertainty dominated by fundamental thermal noise for the first time. We achieve this using the instantaneous velocity of a trapped particle in an optical tweezer. To resolve the instantaneous velocity we develop a structured-light detection system that allows particle tracking with megahertz bandwidths. Our results translate viscosity from a static averaged property, to one that may be dynamically tracked on the timescales of active dynamics. This opens a pathway to new discoveries in out-of-equilibrium systems, from the fast dynamics of phase transitions, to energy dissipation in motor molecule stepping, to violations of fluctuation laws of equilibrium thermodynamics.

cond-mat.soft↗

arXiv2020

Quantum correlations overcome the photodamage limits of light microscopy

State-of-the-art microscopes use intense lasers that can severely disturb biological processes, function and viability. This introduces hard limits on performance that only quantum photon correlations can overcome. Here we demonstrate this absolute quantum advantage, achieving signal-to-noise beyond the photodamage-free capacity of conventional microscopy. We achieve this in a coherent Raman microscope, which we use to image molecular bonds within a cell with both quantum-enhanced contrast and sub-wavelength resolution. This allows the observation of nanoscale biological structures that would otherwise not be resolved. Coherent Raman microscopes allow highly selective biomolecular finger-printing in unlabelled specimens, but photodamage is a major roadblock for many applications. By showing that this roadblock can be overcome, our work provides a path towards order-of-magnitude improvements in both sensitivity and imaging speed.

physics.optics↗