搜索结果：tiny

Detecting objects with visual sensors is crucial for numerous mobile robotics applications, from autonomous navigation to inspection. However, robots often need to operate under significant domains shifts from those they were trained in, requiring them to adjust to these changes. Tiny mobile robots, subject to size, power, and computational constraints, face even greater challenges when running and adapting detection models on low-resolution and noisy images. Such adaptability, though, is crucial for real-world deployment, where robots must operate effectively in dynamic and unpredictable settings. In this work, we introduce a new vision benchmark to evaluate lightweight continual learning strategies tailored to the unique characteristics of tiny robotic platforms. Our contributions include: (i) Tiny Robotics Object Detection~(TiROD), a challenging video dataset collected using the onboard camera of a small mobile robot, designed to test object detectors across various domains and classes; (ii) a comprehensive benchmark of several continual learning strategies on different scenarios using NanoDet, a lightweight, real-time object detector for resource-constrained devices.. Our resul

Tiny face detection aims to find faces with high degrees of variability in scale, resolution and occlusion in cluttered scenes. Due to the very little information available on tiny faces, it is not sufficient to detect them merely based on the information presented inside the tiny bounding boxes or their context. In this paper, we propose to exploit the semantic similarity among all predicted targets in each image to boost current face detectors. To this end, we present a novel framework to model semantic similarity as pairwise constraints within the metric learning scheme, and then refine our predictions with the semantic similarity by utilizing the graph cut techniques. Experiments conducted on three widely-used benchmark datasets have demonstrated the improvement over the-state-of-the-arts gained by applying this idea.

Tiny grains such as PAHs have been thought to dramatically reduce the coupling between gas and magnetic fields in weakly ionized gas such as in protoplanetary disks (PPDs) because they provide tremendous surface area to recombine free electrons. The presence of tiny grains in PPDs thus raises the question of whether the magnetorotational instability (MRI) is able to drive rapid accretion to be consistent with observations. Charged tiny grains have similar conduction properties as ions, whose presence leads to qualitatively new behaviors in the conductivity tensor, characterized by n_bar/n_e>1, where n_e and n_bar denote the number densities of free electrons and all other charged species respectively. In particular, Ohmic conductivity becomes dominated by charged grains rather than electrons when n_bar/n_e exceeds about 10^3, and Hall and ambipolar diffusion (AD) coefficients are reduced by a factor of (n_bar/n_e)^2 in the AD dominated regime relative to that in the Ohmic regime. Applying the methodology of Bai (2011), we find that in PPDs, when PAHs are sufficiently abundant (>1e-9 per H_2), there exists a transition radius r_trans of about 10-20 AU, beyond which the MRI act

Tiny objects, with their limited spatial resolution, often resemble point-like distributions. As a result, bounding box prediction using point-level supervision emerges as a natural and cost-effective alternative to traditional box-level supervision. However, the small scale and lack of distinctive features of tiny objects make point annotations prone to noise, posing significant hurdles for model robustness. To tackle these challenges, we propose Point Teacher--the first end-to-end point-supervised method for robust tiny object detection in aerial images. To handle label noise from scale ambiguity and location shifts in point annotations, Point Teacher employs the teacher-student architecture and decouples the learning into a two-phase denoising process. In this framework, the teacher network progressively denoises the pseudo boxes derived from noisy point annotations, guiding the student network's learning. Specifically, in the first phase, random masking of image regions facilitates regression learning, enabling the teacher to transform noisy point annotations into coarse pseudo boxes. In the second phase, these coarse pseudo boxes are refined using dynamic multiple instance lea

Edges are the fundamental visual element for discovering tiny obstacles using a monocular camera. Nevertheless, tiny obstacles often have weak and inconsistent edge cues due to various properties such as small size and similar appearance to the free space, making it hard to capture them. ...

We report deep, high spectral resolution WSRT HI 21cm observations of four high latitude compact radio sources, that have revealed a new population of tiny, discrete clouds in the diffuse ISM, with peak optical depths tau ~ 0.1-2%, HI column densities of 0.4-8 * 10^{18} cm^{-2} and core temperatures of 20-80 K. Imaging detections confirm these low column densities and imply linear core dimensions of a few thousand AU, assuming a distance of 100 pc. The physical origin of these tiny HI structures and their distribution in the ISM is at present unknown. Further observations will be required to determine whether they are a ubiquitous component of the ISM.

In this paper, we proposed a tiny multi-path CNN-based Super-Resolution (SR) method, called TMSR. We mainly refer to some tiny CNN-based SR methods, under 5k parameters. The main contribution of the proposed method is the improved multi-path learning and self-defined activated function. The experimental results show that TMSR obtains competitive image quality (i.e. PSNR and SSIM) compared to the related works under 5k parameters.

Finding Tiny Faces (by Hu and Ramanan) proposes a novel approach to find small objects in an image. Our contribution consists in deeply understanding the choices of the paper together with applying and extending a similar method to a real world subject which is the counting of people in a public demonstration.

We have implemented support for Padauk microcontrollers, tiny 8-Bit devices with 60 B to 256 B of RAM, in the Small Device C Compiler (SDCC), showing that the use of (mostly) standard C to program such minimal devices is feasible. We report on our experience and on the difficulties in supporting the hardware multithreading present on some of these devices. To make the devices a better target for C, we propose various enhancements of the architecture, and empirically evaluated their impact on code size.

Objects $T$ whose exponential functor $(-)^T$ admits a right adjoint $(-)_T$ are known under different names. The fact that they exist, yet that the only set that satisfies this in the category of sets is the singleton made Lawvere suggest they ought to be ``amazingly tiny'' -- hence Lawvere's acronym ``A.T.O.M.'' This report explores how intuitively tiny any such object is. Evidences both in favor and to the contrary are produced by looking at their categorical behavior (subobjects, quotients, retracts, etc) when the ambient category is a topos. The topological behavior (connectedness, contractibility, connected components, etc) of both $T$ and $(-)_T$ is further analyzed in toposes that satisfy certain precohesive conditions over their decidable objects, where this tininess is tested against parts of Lawvere's foundational proposal for Synthetic Differential Geometry.