arXiv subjects

Janarbek Matai

Publications and source records attributed to Janarbek Matai.

10 recordsLinked to original sources

arXiv2026

Interactive World Simulation with Cosmos for Physical AI

We present a shared framework that adapts Cosmos world foundation models for real-time interactive world simulation in Physical AI. By streaming visual observations in response to actions and scene controls, the framework enables users and embodied agents to interact with evolving virtual environments. It supports both predicting the visual consequences of actions and generating observations from physical state maintained by an external simulator. We demonstrate this approach across camera navigation and robot manipulation, controllable video transfer, and multi-view autonomous vehicle (AV) simulation. These examples provide a shared foundation for developing interactive virtual environments for Physical AI training and evaluation. Our code and model checkpoints are available at https://github.com/nvidia/cosmos and https://huggingface.co/collections/nvidia/cosmos3-examples. The project website is available at https://research.nvidia.com/labs/cosmos-lab/cosmos-simulation.

cs.AI↗

arXiv2026

Thread-Efficient Decoding for Neural Texture Compression

Neural texture compression (NTC) achieves higher compression ratios than BCn formats but suffers from GPU thread divergence, which significantly reduces runtime performance. In this work, we propose a shared decoder MLP architecture -- trained with a gradual decoder freezing schedule -- combined with texture clustering to reduce thread divergence by 25%-52% while preserving rendering quality. We evaluate our method on over 500 textures and multiple real rendering scenes, demonstrating up to 8.48x speedup on the Radeon RX 9070 XT GPU compared to non-shared baselines. Our key contributions include: (1) a unified shared decoder architecture that reduces divergence by grouping textures; (2) a training recipe with gradual decoder freezing that improves stability and reconstruction accuracy; (3) a semantic clustering strategy using CLIP embeddings that groups similar textures for effective decoder sharing; and (4) comprehensive performance and ablation studies validating our approach.

cs.CV↗

arXiv2026

PEPS: Positional Encoding Projected Sampling -- Extended

Implicit neural representations (INRs) are increasingly being used as tools to map coordinates to signals, encompassing applications from neural fields to texture compression, shape representations, and beyond. Most INR methods are based on using high-dimensional projections of the initial coordinates through encoders such as grid or positional encoding. Nevertheless, positional encoding is often insufficient and grids, as we show in this paper, require high resolution for being able to learn. In this paper, we demonstrate that positional encoding can be used not only as a high-dimensional embedding but also decomposed as a series of meaningful points. We propose the Positional Encoding Projected Sampling, where we treat the projection of the original coordinate at each frequency as a point of interest. We describe the motion of each point with respect to the frequencies and show that it follows a unique pattern. Finally, we use the unique motion of each point as a basis decomposition for doing learned positional encoding using grids. We prove, using three competitive applications; image representation, texture compression, and signed distance function; that the proposed approach outperforms the current state of the art methods, and often requires 25\% less parameters for equivalent reconstruction error or rendering.

cs.CV↗

arXiv2023

Restoring the Broken Covenant Between Compilers and Deep Learning Accelerators

Deep learning accelerators address the computational demands of Deep Neural Networks (DNNs), departing from the traditional Von Neumann execution model. They leverage specialized hardware to align with the application domain's structure. Compilers for these accelerators face distinct challenges compared to those for general-purpose processors. These challenges include exposing and managing more micro-architectural features, handling software-managed scratch pads for on-chip storage, explicitly managing data movement, and matching DNN layers with varying hardware capabilities. These complexities necessitate a new approach to compiler design, as traditional compilers mainly focused on generating fine-grained instruction sequences while abstracting micro-architecture details. This paper introduces the Architecture Covenant Graph (ACG), an abstract representation of an architectural structure's components and their programmable capabilities. By enabling the compiler to work with the ACG, it allows for adaptable compilation workflows when making changes to accelerator design, reducing the need for a complete compiler redevelopment. Codelets, which express DNN operation functionality and evolve into execution mappings on the ACG, are key to this process. The Covenant compiler efficiently targets diverse deep learning accelerators, achieving 93.8% performance compared to state-of-the-art, hand-tuned DNN layer implementations when compiling 14 DNN layers from various models on two different architectures.

cs.DC↗

arXiv2023

Factorized Inverse Path Tracing for Efficient and Accurate Material-Lighting Estimation

Inverse path tracing has recently been applied to joint material and lighting estimation, given geometry and multi-view HDR observations of an indoor scene. However, it has two major limitations: path tracing is expensive to compute, and ambiguities exist between reflection and emission. Our Factorized Inverse Path Tracing (FIPT) addresses these challenges by using a factored light transport formulation and finds emitters driven by rendering errors. Our algorithm enables accurate material and lighting optimization faster than previous work, and is more effective at resolving ambiguities. The exhaustive experiments on synthetic scenes show that our method (1) outperforms state-of-the-art indoor inverse rendering and relighting methods particularly in the presence of complex illumination effects; (2) speeds up inverse path tracing optimization to less than an hour. We further demonstrate robustness to noisy inputs through material and lighting estimates that allow plausible relighting in a real scene. The source code is available at: https://github.com/lwwu2/fipt

cs.CV↗

arXiv2023

Comprehensive Benchmarking of Binary Neural Networks on NVM Crossbar Architectures

Non-volatile memory (NVM) crossbars have been identified as a promising technology, for accelerating important machine learning operations, with matrix-vector multiplication being a key example. Binary neural networks (BNNs) are especially well-suited for use with NVM crossbars due to their use of a low-bitwidth representation for both activations and weights. However, the aggressive quantization of BNNs can result in suboptimal accuracy, and the analog effects of NVM crossbars can further degrade the accuracy during inference. This paper presents a comprehensive study that benchmarks BNNs trained and validated on ImageNet and deployed on NeuroSim, a simulator for NVM-crossbar-based PIM architecture. Our study analyzes the impact of various parameters, such as input precision and ADC resolution, on both the accuracy of the inference and the hardware performance metrics. We have found that an ADC resolution of 8-bit with an input precision of 4-bit achieves near-optimal accuracy compared to the original BNNs. In addition, we have identified bottleneck components in the PIM architecture that affect area, latency, and energy consumption, and we demonstrate the impact that different BNN layers have on hardware performance.

cs.ET↗

arXiv2022

IRISformer: Dense Vision Transformers for Single-Image Inverse Rendering in Indoor Scenes

Indoor scenes exhibit significant appearance variations due to myriad interactions between arbitrarily diverse object shapes, spatially-changing materials, and complex lighting. Shadows, highlights, and inter-reflections caused by visible and invisible light sources require reasoning about long-range interactions for inverse rendering, which seeks to recover the components of image formation, namely, shape, material, and lighting. In this work, our intuition is that the long-range attention learned by transformer architectures is ideally suited to solve longstanding challenges in single-image inverse rendering. We demonstrate with a specific instantiation of a dense vision transformer, IRISformer, that excels at both single-task and multi-task reasoning required for inverse rendering. Specifically, we propose a transformer architecture to simultaneously estimate depths, normals, spatially-varying albedo, roughness and lighting from a single image of an indoor scene. Our extensive evaluations on benchmark datasets demonstrate state-of-the-art results on each of the above tasks, enabling applications like object insertion and material editing in a single unconstrained real image, with greater photorealism than prior works. Code and data are publicly released at https://github.com/ViLab-UCSD/IRISformer.

cs.CV↗

arXiv2021

X-Distill: Improving Self-Supervised Monocular Depth via Cross-Task Distillation

In this paper, we propose a novel method, X-Distill, to improve the self-supervised training of monocular depth via cross-task knowledge distillation from semantic segmentation to depth estimation. More specifically, during training, we utilize a pretrained semantic segmentation teacher network and transfer its semantic knowledge to the depth network. In order to enable such knowledge distillation across two different visual tasks, we introduce a small, trainable network that translates the predicted depth map to a semantic segmentation map, which can then be supervised by the teacher network. In this way, this small network enables the backpropagation from the semantic segmentation teacher's supervision to the depth network during training. In addition, since the commonly used object classes in semantic segmentation are not directly transferable to depth, we study the visual and geometric characteristics of the objects and design a new way of grouping them that can be shared by both tasks. It is noteworthy that our approach only modifies the training process and does not incur additional computation during inference. We extensively evaluate the efficacy of our proposed approach on the standard KITTI benchmark and compare it with the latest state of the art. We further test the generalizability of our approach on Make3D. Overall, the results show that our approach significantly improves the depth estimation accuracy and outperforms the state of the art.

cs.CV↗

arXiv2018

Parallel Programming for FPGAs

This book focuses on the use of algorithmic high-level synthesis (HLS) to build application-specific FPGA systems. Our goal is to give the reader an appreciation of the process of creating an optimized hardware design using HLS. Although the details are, of necessity, different from parallel programming for multicore processors or GPUs, many of the fundamental concepts are similar. For example, designers must understand memory hierarchy and bandwidth, spatial and temporal locality of reference, parallelism, and tradeoffs between computation and storage. This book is a practical guide for anyone interested in building FPGA systems. In a university environment, it is appropriate for advanced undergraduate and graduate courses. At the same time, it is also useful for practicing system designers and embedded programmers. The book assumes the reader has a working knowledge of C/C++ and includes a significant amount of sample code. In addition, we assume familiarity with basic computer architecture concepts (pipelining, speedup, Amdahl's Law, etc.). A knowledge of the RTL-based FPGA design flow is helpful, although not required.

cs.AR↗

arXiv2014

Enabling FPGAs for the Masses

Implementing an application on a FPGA remains a difficult, non-intuitive task that often requires hardware design expertise in a hardware description language (HDL). High-level synthesis (HLS) raises the design abstraction from HDL to languages such as C/C++/Scala/Java. Despite this, in order to get a good quality of result (QoR), a designer must carefully craft the HLS code. In other words, HLS designers must implement the application using an abstract language in a manner that generates an efficient micro-architecture; we call this process writing restructured code. This reduces the benefits of implementing the application at a higher level of abstraction and limits the impact of HLS by requiring explicit knowledge of the underlying hardware architecture. Developers must know how to write code that reflects low level implementation details of the application at hand as it is interpreted by HLS tools. As a result, FPGA design still largely remains job of either hardware engineers or expert HLS designers. In this work, we aim to take a step towards making HLS tools useful for a broader set of programmers. To do this, we study methodologies of restructuring software code for HLS tools; we provide examples of designing different kernels in state-of-the art HLS tools; and we present a list of challenges for developing a hardware programming model for software programmers.

cs.SE↗