Search for: All records

While the shape of the Lyα profile is viewed as one of the best tracers of ionizing-photon escape fraction (fesc) within low redshift (z~0.3) surveys of the Lyman continuum, this connection remains untested at high redshift. Here, we combine deep, rest-UV Keck/LRIS spectra of 80 objects from the Keck Lyman Continuum Spectroscopic Survey with rest-optical Keck/MOSFIRE spectroscopy in order to examine potential correlations between Lyα profile shape and the escape of ionizing radiation within z~3 star-forming galaxies. We measure the velocity separation between double-peaked Lyα emission structure (vsep), between red-side Lyα emission peaks and systemic (vLyα,red), and between red-side emission peaks and low-ionization interstellar absorption lines (vLyα−LIS). We find that the IGM-corrected ratio of ionizing to non-ionizing flux density is significantly higher in KLCS objects with lower vLyα,red. We find no significant trend between measures of ionizing-photon escape and vLyα−LIS. We compare our results to measurements of z~0.3 "Green Peas" from the literature and find that KLCS objects have larger vsep at fixed vLyα,red, larger fesc at fixed vLyα,red, and higher vLyα,red overall than z~0.3 analogs. We conclude that the Lyα profile shapes of our high-redshift sources are fundamentally different, and that measurements of profile shape such as vLyα,red map on to fesc in different ways. We caution against building reionization-era fesc diagnostics based purely on Lyα profiles of low-redshift dwarf galaxies. Tracing vsep, vLyα,red, and fesc in a larger sample of z~3 galaxies will reveal how these variables may be connected for galaxies at the epoch of reionization. 

Reconstructing 4D vehicular activity (3D space and time) from cameras is useful for autonomous vehicles, commuters and local authorities to plan for smarter and safer cities. Traffic is inherently repetitious over long periods, yet current deep learning-based 3D reconstruction methods have not considered such repetitions and have difficulty generalizing to new intersection-installed cameras. We present a novel approach exploiting longitudinal (long-term) repetitious motion as self-supervision to reconstruct 3D vehicular activity from a video captured by a single fixed camera. Starting from off-the-shelf 2D keypoint detections, our algorithm optimizes 3D vehicle shapes and poses, and then clusters their trajectories in 3D space. The 2D keypoints and trajectory clusters accumulated over long-term are later used to improve the 2D and 3D keypoints via self-supervision without any human annotation. Our method improves reconstruction accuracy over state of the art on scenes with a significant visual difference from the keypoint detector’s training data, and has many applications including velocity estimation, anomaly detection and vehicle counting. We demonstrate results on traffic videos captured at multiple city intersections, collected using our smartphones, YouTube, and other public datasets. 

We consider the task of 3D pose estimation and tracking of multiple people seen in an arbitrary number of camera feeds. We propose TesseTrack, a novel top-down approach that simultaneously reasons about multiple individuals’ 3D body joint reconstructions and associations in space and time in a single end-to-end learnable framework. At the core of our approach is a novel spatio-temporal formulation that operates in a common voxelized feature space aggregated from single- or multiple camera views. After a person detection step, a 4D CNN produces short-term persons pecific representations which are then linked across time by a differentiable matcher. The linked descriptions are then merged and deconvolved into 3D poses. This joint spatio-temporal formulation contrasts with previous piecewise strategies that treat 2D pose estimation, 2D-to-3D lifting, and 3D pose tracking as independent sub-problems that are error-prone when solved in isolation. Furthermore, unlike previous methods, TesseTrack is robust to changes in the number of camera views and achieves very good results even if a single view is available at inference time. Quantitative evaluation of 3D pose reconstruction accuracy on standard benchmarks shows significant improvements over the state of the art. Evaluation of multi-person articulated 3D pose tracking in our novel evaluation framework demonstrates the superiority of TesseTrack over strong baselines. 

ABSTRACT We present results from the NIRVANDELS survey on the gas-phase metallicity (Zg, tracing O/H) and stellar metallicity (Z⋆, tracing Fe/H) of 33 star-forming galaxies at redshifts 2.95 < z < 3.80. Based on a combined analysis of deep optical and near-IR spectra, tracing the rest-frame far-ultraviolet (FUV; 1200–2000 Å) and rest-frame optical (3400–5500 Å), respectively, we present the first simultaneous determination of the stellar and gas-phase mass–metallicity relationships (MZRs) at z ≃ 3.4. In both cases, we find that metallicity increases with increasing stellar mass (M⋆) and that the power-law slope at M⋆ ≲ 1010M⊙ of both MZRs scales as $$Z \propto M_{\star }^{0.3}$$. Comparing the stellar and gas-phase MZRs, we present direct evidence for super-solar O/Fe ratios (i.e. α-enhancement) at z > 3, finding (O/Fe) = 2.54 ± 0.38 × (O/Fe)⊙, with no clear dependence on M⋆.