no code implementations • 10 Apr 2024 • Oğuzhan Fatih Kar, Alessio Tonioni, Petra Poklukar, Achin Kulshrestha, Amir Zamir, Federico Tombari
Our results highlight the potential of incorporating different visual biases for a more broad and contextualized visual understanding of VLMs.
no code implementations • 22 Mar 2024 • Teresa Yeo, Andrei Atanov, Harold Benoit, Aleksandr Alekseev, Ruchira Ray, Pooya Esmaeil Akhoondi, Amir Zamir
In this work, we present a method to control a text-to-image generative model to produce training data specifically "useful" for supervised learning.
no code implementations • 26 Dec 2023 • Harold Benoit, Liangze Jiang, Andrei Atanov, Oğuzhan Fatih Kar, Mattia Rigotti, Amir Zamir
We show that (1) diversification methods are highly sensitive to the distribution of the unlabeled data used for diversification and can underperform significantly when away from a method-specific sweet spot.
no code implementations • NeurIPS 2023 • David Mizrahi, Roman Bachmann, Oğuzhan Fatih Kar, Teresa Yeo, Mingfei Gao, Afshin Dehghan, Amir Zamir
Current machine learning models for vision are often highly specialized and limited to a single modality and task.
no code implementations • ICCV 2023 • Teresa Yeo, Oğuzhan Fatih Kar, Zahra Sodagar, Amir Zamir
We propose a method for adapting neural networks to distribution shifts at test-time.
1 code implementation • CVPR 2023 • Marius Memmel, Roman Bachmann, Amir Zamir
Our model outperforms previous methods while training on only a fraction of the data.
no code implementations • 20 Dec 2022 • Yajie Bao, Yang Li, Shao-Lun Huang, Lin Zhang, Lizhong Zheng, Amir Zamir, Leonidas Guibas
Task transfer learning is a popular technique in image processing applications that uses pre-trained models to reduce the supervision cost of related tasks.
no code implementations • 8 Dec 2022 • Onur Beker, Mohammad Mohammadi, Amir Zamir
For training these perceptual representations, we combine Q-learning with contrastive representation learning to create a latent space where the distance between the embeddings of two states captures how easily an optimal policy can traverse between them.
no code implementations • 1 Dec 2022 • Andrei Atanov, Andrei Filatov, Teresa Yeo, Ajay Sohmshetty, Amir Zamir
An intriguing question would be: what if, instead of fixing the task and searching in the model space, we fix the model and search in the task space?
1 code implementation • 4 Apr 2022 • Roman Bachmann, David Mizrahi, Andrei Atanov, Amir Zamir
We show this pre-training strategy leads to a flexible, simple, and efficient framework with improved transfer results to downstream tasks.
Ranked #1 on Semantic Segmentation on Hypersim
1 code implementation • CVPR 2022 • Oğuzhan Fatih Kar, Teresa Yeo, Andrei Atanov, Amir Zamir
We introduce a set of image transformations that can be used as corruptions to evaluate the robustness of models as well as data augmentation mechanisms for training neural networks.
1 code implementation • 11 Feb 2022 • Yael Vinker, Ehsan Pajouheshgar, Jessica Y. Bo, Roman Christian Bachmann, Amit Haim Bermano, Daniel Cohen-Or, Amir Zamir, Ariel Shamir
Abstraction is at the heart of sketching due to the simple and minimal nature of line drawings.
no code implementations • 7 Feb 2022 • Andrei Atanov, Shijian Xu, Onur Beker, Andrei Filatov, Amir Zamir
Transfer learning has witnessed remarkable progress in recent years, for example, with the introduction of augmentation-based contrastive self-supervised learning methods.
1 code implementation • ICCV 2021 • Ainaz Eftekhar, Alexander Sax, Roman Bachmann, Jitendra Malik, Amir Zamir
This paper introduces a pipeline to parametrically sample and render multi-task vision datasets from comprehensive 3D scans from the real world.
no code implementations • 29 Sep 2021 • Andrei Atanov, Shijian Xu, Onur Beker, Andrey Filatov, Amir Zamir
Self-supervised learning has witnessed remarkable progress in recent years, in particular with the introduction of augmentation-based contrastive methods.
no code implementations • ICCV 2021 • Teresa Yeo, Oğuzhan Fatih Kar, Alexander Sax, Amir Zamir
We present a method for making neural network predictions robust to shifts from the training data distribution.
no code implementations • 1 Jan 2021 • Teresa Yeo, Oguzhan Fatih Kar, Amir Zamir
We present a method for making predictions using neural networks that, at the test time, is robust against shifts from the training data distribution.
no code implementations • 13 Nov 2020 • Bryan Chen, Alexander Sax, Gene Lewis, Iro Armeni, Silvio Savarese, Amir Zamir, Jitendra Malik, Lerrel Pinto
Vision-based robotics often separates the control loop into one module for perception and a separate module for control.
1 code implementation • 7 Jun 2020 • Amir Zamir, Alexander Sax, Teresa Yeo, Oğuzhan Kar, Nikhil Cheerla, Rohan Suri, Zhangjie Cao, Jitendra Malik, Leonidas Guibas
Visual perception entails solving a wide set of tasks, e. g., object detection, depth estimation, etc.
2 code implementations • ECCV 2020 • Jeffrey O. Zhang, Alexander Sax, Amir Zamir, Leonidas Guibas, Jitendra Malik
When training a neural network for a desired task, one may prefer to adapt a pre-trained network rather than starting from randomly initialized weights.
1 code implementation • 23 Dec 2019 • Alexander Sax, Jeffrey O. Zhang, Bradley Emi, Amir Zamir, Silvio Savarese, Leonidas Guibas, Jitendra Malik
How much does having visual priors about the world (e. g. the fact that the world is 3D) assist in learning to perform downstream motor tasks (e. g. navigating a complex environment)?
5 code implementations • CVPR 2018 • Fei Xia, Amir Zamir, Zhi-Yang He, Alexander Sax, Jitendra Malik, Silvio Savarese
Developing visual perception models for active agents and sensorimotor control are cumbersome to be done in the physical world, as existing algorithms are too slow to efficiently learn in real-time and robots are fragile and costly.
1 code implementation • CVPR 2018 • Amir Zamir, Alexander Sax, William Shen, Leonidas Guibas, Jitendra Malik, Silvio Savarese
The product is a computational taxonomic map for task transfer learning.
no code implementations • 31 Oct 2015 • Francesco Castaldo, Amir Zamir, Roland Angst, Francesco Palmieri, Silvio Savarese
In this paper, we therefore explore this idea and propose an automatic method for detecting and representing the semantic information of an RGB image with the goal of performing cross-view matching with a (non-RGB) geographic information system (GIS).
no code implementations • ICCV 2015 • Ozan Sener, Amir Zamir, Silvio Savarese, Ashutosh Saxena
The proposed method is capable of providing a semantic "storyline" of the video composed of its objective steps.