One-shot channel simulation has recently emerged as a promising alternative to quantization and entropy coding in machine-learning-based lossy data compression schemes. How-ever, while there are several potential applications of channel simulation - lossy compression with realism constraints or differential privacy, to name a few - little is known about its fundamental limitations. In this paper, we restrict our attention to a subclass of channel simulation protocols called causal rejection samplers (CRS), establish new, tighter lower bounds on their expected runtime and codelength, and demonstrate the bounds' achievability. Concretely, for an arbitrary CRS, let$Q$and$P$denote a target and proposal distribution supplied as input, and let$K$be the number of samples examined by the algorithm. We show that the expected runtime$\mathbb{E}[K]$of any CRS scales at least as$\exp_{2}(D_{\infty}[Q\Vert P])$, where$D_{\infty}[Q\Vert P]$is the Renyi$\infty$-divergence. Regarding the codelength, we show that$D_{\text{KL}}[Q\Vert P]\leq D_{CS}[Q\Vert P]\leq \mathbb{H}[K]$, where$D_{CS}[Q\Vert P]$is a new quantity we call the channel simulation divergence. Furthermore, we prove that our new lower bound, unlike the$D_{\text{KL}}[Q\Vert P]$lower bound, is achievable tightly, i.e. there is a CRS such that$\mathbb{H}[K]\leq D_{CS}[Q\Vert P]+\log_{2}(e+1)$. Finally, we conduct numerical studies of the asymptotic scaling of the codelength of Gaussian and Laplace channel simulation algorithms.

For telepresence to support the richness of multiparty conversations, it is important to convey motion parallax and stereoscopy without head-worn apparatus. TeleHuman2 is a "hologrammatic" telepresence system that conveys full-body 3D video of interlocutors using a human-sized cylindrical light field display. For rendering, the system uses an array of projectors mounted above the heads of participants in a ring around a retroreflective cylinder. Unique angular renditions are calculated from streaming depth video captured at the remote location. Projected images are retro-reflected into the eyes of local participants, at 1.3º intervals providing angular renditions simultaneously for left and right eyes of all onlookers, which conveys motion parallax and stereoscopy without head-worn apparatus or head tracking. Our technical evaluation of the angular accuracy of the system demonstrates that the error in judging the angle of a remote arrow object represented in TeleHuman2 is within 1 degree, and not significantly different from similar judgments of a collocated arrow object.

We present HoloFlex, a 3D flexible smartphone featuring a light-field display consisting of a high-resolution P-OLED display and an array of 16,640 microlenses. HoloFlex allows mobile users to interact with 3D images featuring natural visual cues such as motion parallax and stereoscopy without glasses or head tracking. Its flexibility allows the use of bend input for interacting with 3D objects along the z axis. Images are rendered into 12-pixel wide circular blocks-pinhole views of the 3D scene-which enable ~80 unique viewports at an effective resolution of 160 × 104. The microlens array distributes each pixel from the display in a direction that preserves the angular information of light rays in the 3D scene. We present a preliminary study evaluating the effect of bend input vs. a vertical touch screen slider on 3D docking performance. Results indicate that bend input significantly improves movement time in this task. We also present 3D applications including a 3D editor, a 3D Angry Birds game and a 3D teleconferencing system that utilize bend input.