Zobrazeno 1 - 10
of 47 728
pro vyhledávání: '"Shrivastava, A."'
Autor:
Saini, Nirat, Bodla, Navaneeth, Shrivastava, Ashish, Ravichandran, Avinash, Zhang, Xiao, Shrivastava, Abhinav, Singh, Bharat
We introduce InVi, an approach for inserting or replacing objects within videos (referred to as inpainting) using off-the-shelf, text-to-image latent diffusion models. InVi targets controlled manipulation of objects and blending them seamlessly into
Externí odkaz:
http://arxiv.org/abs/2407.10958
Implicit Neural Networks (INRs) have emerged as powerful representations to encode all forms of data, including images, videos, audios, and scenes. With video, many INRs for video have been proposed for the compression task, and recent methods featur
Externí odkaz:
http://arxiv.org/abs/2408.02672
Prior art in traffic incident detection relies on high sensor coverage and is primarily based on decision-tree and random forest models that have limited representation capacity and, as a result, cannot detect incidents with high accuracy. This paper
Externí odkaz:
http://arxiv.org/abs/2408.00996
Autor:
Lin, Xi Victoria, Shrivastava, Akshat, Luo, Liang, Iyer, Srinivasan, Lewis, Mike, Gosh, Gargi, Zettlemoyer, Luke, Aghajanyan, Armen
We introduce MoMa, a novel modality-aware mixture-of-experts (MoE) architecture designed for pre-training mixed-modal, early-fusion language models. MoMa processes images and text in arbitrary sequences by dividing expert modules into modality-specif
Externí odkaz:
http://arxiv.org/abs/2407.21770
We propose a simple yet effective approach for few-shot action recognition, emphasizing the disentanglement of motion and appearance representations. By harnessing recent progress in tracking, specifically point trajectories and self-supervised repre
Externí odkaz:
http://arxiv.org/abs/2407.18249
We propose WayEx, a new method for learning complex goal-conditioned robotics tasks from a single demonstration. Our approach distinguishes itself from existing imitation learning methods by demanding fewer expert examples and eliminating the need fo
Externí odkaz:
http://arxiv.org/abs/2407.15849
Trusted execution environments (TEEs) are an integral part of modern secure processors. They ensure that their application and code pages are confidential, tamper proof and immune to diverse types of attacks. In 2021, Intel suddenly announced its pla
Externí odkaz:
http://arxiv.org/abs/2407.13572
Securing deep neural networks (DNNs) from side-channel attacks is an important problem as of today, given the substantial investment of time and resources in acquiring the raw data and training complex models. All published countermeasures (CMs) add
Externí odkaz:
http://arxiv.org/abs/2407.13383
Autor:
Zhang, Tianyi, Shrivastava, Anshumali
Large language models (LLMs) have numerous applications across various domains, but their high computational and memory demands pose significant deployment challenges. Weight quantization is an effective technique for reducing the decoding latency an
Externí odkaz:
http://arxiv.org/abs/2407.10032
Autor:
Levy, Mara, Shrivastava, Abhinav
Learning to represent three dimensional (3D) human pose given a two dimensional (2D) image of a person, is a challenging problem. In order to make the problem less ambiguous it has become common practice to estimate 3D pose in the camera coordinate s
Externí odkaz:
http://arxiv.org/abs/2407.07092