Search

Manually creating textures for 3D meshes is time-consuming, even for expert visual content creators. We propose a fast approach for automatically texturing an input 3D mesh based on a user-provided text prompt. Importantly, our approach disentangles lighting from surface material/reflectance in the resulting texture so that the mesh can be properly relit and rendered in any lighting environment. We introduce LightControlNet, a new text-to-image model based on the ControlNet architecture, which allows the specification of the desired lighting as a conditioning image to the model. Our text-to-texture pipeline then constructs the texture in two stages. The first stage produces a sparse set of visually consistent reference views of the mesh using LightControlNet. The second stage applies a texture optimization based on Score Distillation Sampling (SDS) that works with LightControlNet to increase the texture quality while disentangling surface material from lighting. Our algorithm is significantly faster than previous text-to-texture methods, while producing high-quality and relightable textures., Comment: Project page: https://flashtex.github.io/

The intermittency phenomenon is the occurrence of very high but rare peaks, which despite their rarity influence the asymptotic behaviour of the underlying system. Mathematically this can be characterised with the asymptotics of moments. In this article we show the existence of intermittency phenomena for SDEs with interaction with dissipative coefficients by showing uniform convergence of their Lyapunov exponents.

Understanding fault types can lead to novel approaches to debugging and runtime verification. Dealing with complex faults, particularly in the challenging area of embedded systems, craves for more powerful tools, which are now becoming available to engineers.

Content-Addressable Memory (CAM) is a powerful abstraction for building memory caches, routing tables and hazard detection logic. Without a native CAM structure available on FPGA devices, their functionality must be emulated using the structural primitives at hand. Such an emulation causes significant overhead in the consumption of the underlying resources, typically general-purpose fabric and on-chip block RAM (BRAM). This often motivates mitigating trade-offs, such as the reduction of the associativity of memory caches. This paper describes a technique to implement the hazard resolution in a memory update queue that hides the off-chip memory readout latency of read-modify-write cycles while guaranteeing the delivery of the full memory bandwidth. The innovative use of DSP slices allows them to assume and combine the functions of (a) the tag and data storage, (b) the tag matching, and (c) the data update in this key-value storage scenario. The proposed approach provides designers with extra flexibility by adding this resource type as another option to implement CAM., Comment: Submitted to FPL 2020

Runtime verification is an area of formal methods that studies the dynamic analysis of execution traces against formal specifications. Typically, the two main activities in runtime verification efforts are the process of creating monitors from specifications, and the algorithms for the evaluation of traces against the generated monitors. Other activities involve the instrumentation of the system to generate the trace and the communication between the system under analysis and the monitor. Most of the applications in runtime verification have been focused on the dynamic analysis of software, even though there are many more potential applications to other computational devices and target systems. In this paper we present a collection of challenges for runtime verification extracted from concrete application domains, focusing on the difficulties that must be overcome to tackle these specific challenges. The computational models that characterize these domains require to devise new techniques beyond the current state of the art in runtime verification.

In this paper, we revisit scalar field theories in $d$ space-time dimensions possessing $U(N)$ global symmetry. Following our recent work arXiv:1402.1430v2, we consider the generating function of correlation functions of all $U(N)$-invariant, single-trace operators at the free fixed point. The exact renormalization group equations are cast as Hamilton equations of radial evolution in a model space-time of one higher dimension, in this case $AdS_{d+1}$. The geometry associated with the RG equations is seen to emerge naturally out of the infinite jet bundle corresponding to the field theory, and suggests their interpretation as higher-spin equations of motion. While the higher-spin equations we obtain are remarkably simple, they are non-local in an essential way. Nevertheless, solving these bulk equations of motion in terms of a boundary source, we derive the on-shell action and demonstrate that it correctly encodes all of the correlation functions of the field theory, written as `Witten diagrams'. Since the model space-time has the isometries of the fixed point, it is possible to construct new higher spin theories defined in terms of geometric structures over other model space-times. We illustrate this by explicitly constructing the higher spin RG equations corresponding to the $z=2$ non-relativistic free field theory in $D$ spatial dimensions. In this case, the model space-time is the Schr\"odinger space-time, $Schr_{D+3}$., Comment: 37 pages, 3 figures; v2: Typos fixed, added discussion about boundary conditions

We consider the Wilson-Polchinski exact renormalization group applied to the generating functional of single-trace operators at a free-fixed point in $d=2+1$ dimensions. By exploiting the rich symmetry structure of free field theory, we study the geometric nature of the RG equations and the associated Ward identities. The geometry, as expected, is holographic, with $AdS$ spacetime emerging correspondent with RG fixed points. The field theory construction gives us a particular vector bundle over the $d+1$-dimensional RG mapping space, called a jet bundle, whose structure group arises from the linear orthogonal bi-local transformations of the bare fields in the path integral. The sources for quadratic operators constitute a connection on this bundle and a section of its endomorphism bundle. Recasting the geometry in terms of the corresponding principal bundle, we arrive at a structure remarkably similar to the Vasiliev theory, where the horizontal part of the connection on the principal bundle is Vasiliev's higher spin connection, while the vertical part (the Faddeev-Popov ghost) corresponds to the $S$-field. The Vasiliev equations are then, respectively, the RG equations and the BRST equations, with the RG beta functions encoding bulk interactions. Finally, we remark that a large class of interacting field theories can be studied through integral transforms of our results, and it is natural to organize this in terms of a large $N$ expansion., Comment: 44 pages, 4 figures; v2 contains additional references, elimination of typos and tightening up of constructions