no code implementations • 3 Jan 2024 • Kyle Buettner, Sina Malakouti, Xiang Lorraine Li, Adriana Kovashka
In the absence of training data from target geographies, we hypothesize that geographically diverse descriptive knowledge of categories can enhance robustness.
1 code implementation • BMVC 2023 • Sina Malakouti, Adriana Kovashka
Existing domain adaptation (DA) and generalization (DG) methods in object detection enforce feature alignment in the visual space but face challenges like object appearance variability and scene complexity, which make it difficult to distinguish between objects and achieve accurate detection.
Ranked #1 on Object Detection on PASCAL VOC to Comic2k
no code implementations • 11 Jun 2023 • Yuya Asano, Diane Litman, Mingzhi Yu, Nikki Lobczowski, Timothy Nokes-Malach, Adriana Kovashka, Erin Walker
While speech-enabled teachable agents have some advantages over typing-based ones, they are vulnerable to errors stemming from misrecognition by automatic speech recognition (ASR).
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 25 Apr 2023 • Giacomo Nebbia, Adriana Kovashka
In this work, we investigate hypernymization as a way to deal with named entities for pretraining grounding-based multi-modal models and for fine-tuning on open-vocabulary detection.
no code implementations • 20 Mar 2023 • Cagri Gungor, Adriana Kovashka
We propose an amplifier method for enhancing the performance of WSOD by integrating depth information.
no code implementations • 17 Mar 2023 • Kyle Buettner, Adriana Kovashka
Methods are mostly evaluated in terms of how well object class names are learned, but captions also contain rich attribute context that should be considered when learning object alignment.
no code implementations • 16 Mar 2023 • Arushi Rai, Adriana Kovashka
The use of large-scale vision-language datasets is limited for object detection due to the negative impact of label noise on localization.
no code implementations • 9 Mar 2023 • Mesut Erhan Unal, Adriana Kovashka
In this paper, we tackle HOI detection with the weakest supervision setting in the literature, using only image-level interaction labels, with the help of a pretrained vision-language model (VLM) and a large language model (LLM).
no code implementations • 9 Dec 2022 • Kyle Buettner, Adriana Kovashka
To address this gap, we conduct an empirical study of contrastive learning and out-of-domain object detection, studying how contrastive view design affects robustness.
no code implementations • SIGDIAL (ACL) 2022 • Yuya Asano, Diane Litman, Mingzhi Yu, Nikki Lobczowski, Timothy Nokes-Malach, Adriana Kovashka, Erin Walker
Speakers build rapport in the process of aligning conversational behaviors with each other.
no code implementations • IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2022 • Christopher Thomas, Adriana Kovashka
Existing cross-modal retrieval methods assume a straightforward relationship where images and text contain portrayals or mentions of the same objects.
Ranked #1 on Cross-Modal Retrieval on COCO 2014 (using extra training data)
1 code implementation • 10 Jun 2022 • Nasrin Kalanat, Adriana Kovashka
In this paper, we use a scene graph, a graph representation of an image, to capture visual components.
no code implementations • 12 May 2022 • Keren Ye, Adriana Kovashka
We explored how to eliminate the expensive annotations in video detection data which provide refined boundaries.
no code implementations • 29 Sep 2021 • Mesut Erhan Unal, Adriana Kovashka
We present a framework to better leverage natural language supervision for a specific downstream task, namely weakly-supervised object detection (WSOD).
1 code implementation • 24 Jun 2021 • Katelyn Morrison, Benjamin Gilby, Colton Lipchak, Adam Mattioli, Adriana Kovashka
We find that vision transformer architectures are inherently more robust to corruptions than the ResNet-50 and MLP-Mixers.
1 code implementation • CVPR 2021 • Keren Ye, Adriana Kovashka
Prior work in scene graph generation requires categorical supervision at the level of triplets - subjects and objects, and predicates that relate them, either with or without bounding box information.
no code implementations • 7 May 2021 • Mingda Zhang, Chun-Te Chu, Andrey Zhmoginov, Andrew Howard, Brendan Jou, Yukun Zhu, Li Zhang, Rebecca Hwa, Adriana Kovashka
With early termination, the average cost can be further reduced to 198M MAdds while maintaining accuracy of 80. 0% on ImageNet.
Ranked #664 on Image Classification on ImageNet
no code implementations • CVPR 2021 • Mingda Zhang, Tristan Maidment, Ahmad Diab, Adriana Kovashka, Rebecca Hwa
The observation that computer vision methods overfit to dataset specifics has inspired diverse attempts to make object recognition models robust to domain shifts.
no code implementations • 4 Jan 2021 • Keren Ye, Adriana Kovashka, Mark Sandler, Menglong Zhu, Andrew Howard, Marco Fornoni
In this paper we address the question: can task-specific detectors be trained and represented as a shared set of weights, plus a very small set of additional weights for each task?
1 code implementation • ICCV 2021 • Meiqi Guo, Rebecca Hwa, Adriana Kovashka
We propose a new approach to detect atypicality in persuasive imagery.
no code implementations • 3 Dec 2020 • Christopher Thomas, Yale Song, Adriana Kovashka
We study the problem of animating images by transferring spatio-temporal visual effects (such as melting) from a collection of videos.
no code implementations • ECCV 2020 • Christopher Thomas, Adriana Kovashka
The abundance of multimodal data (e. g. social media posts) has inspired interest in cross-modal retrieval methods.
1 code implementation • NeurIPS 2019 • Christopher Thomas, Adriana Kovashka
We collect a dataset of over one million unique images and associated news articles from left- and right-leaning news sources, and develop a method to predict the image's political leaning.
1 code implementation • ICCV 2019 • Keren Ye, Mingda Zhang, Adriana Kovashka, Wei Li, Danfeng Qin, Jesse Berent
Learning to localize and name object instances is a fundamental problem in vision, but state-of-the-art approaches rely on expensive bounding box supervision.
no code implementations • 15 Jan 2019 • James Hahn, Adriana Kovashka
They are attractive to companies and nearly unavoidable for consumers.
no code implementations • 28 Dec 2018 • Christopher Thomas, Adriana Kovashka
To do so, we introduce a complementary training modality constructed to be similar in artistic style to the target domain, and enforce that the network learns features that are invariant between the two training modalities.
no code implementations • 25 Nov 2018 • Keren Ye, Mingda Zhang, Wei Li, Danfeng Qin, Adriana Kovashka, Jesse Berent
To alleviate the cost of obtaining accurate bounding boxes for training today's state-of-the-art object detection models, recent weakly supervised detection work has proposed techniques to learn from image-level labels.
no code implementations • 29 Jul 2018 • Keren Ye, Kyle Buettner, Adriana Kovashka
We dedicate our study to understand the dynamic structure of video ads automatically.
no code implementations • 25 Jul 2018 • Christopher Thomas, Adriana Kovashka
We show how our model can be used to produce visually distinct faces which appear to be from a fixed ad topic category.
no code implementations • 21 Jul 2018 • Mingda Zhang, Rebecca Hwa, Adriana Kovashka
Images and text in advertisements interact in complex, non-literal ways.
no code implementations • 8 May 2018 • Nils Murrugarra-Llerena, Adriana Kovashka
How would you search for a unique, fashionable shoe that a friend wore and you want to buy, but you didn't take a picture?
no code implementations • ECCV 2018 • Keren Ye, Adriana Kovashka
In order to convey the most content in their limited space, advertisements embed references to outside knowledge via symbolism.
no code implementations • CVPR 2017 • Zaeem Hussain, Mingda Zhang, Xiaozhong Zhang, Keren Ye, Christopher Thomas, Zuha Agha, Nathan Ong, Adriana Kovashka
There is more to images than their objective physical content: for example, advertisements are created to persuade a viewer to take a certain action.
no code implementations • 7 Nov 2016 • Adriana Kovashka, Olga Russakovsky, Li Fei-Fei, Kristen Grauman
Computer vision systems require large amounts of manually annotated data to properly learn challenging visual concepts.
no code implementations • CVPR 2016 • Christopher Thomas, Adriana Kovashka
To explore the feasibility of current computer vision techniques to address this problem, we created a new dataset of over 180, 000 images taken by 41 well-known photographers.
no code implementations • 15 May 2015 • Adriana Kovashka, Devi Parikh, Kristen Grauman
We propose a novel mode of feedback for image search, where a user describes which properties of exemplar images should be adjusted in order to more closely match his/her mental model of the image sought.
no code implementations • 15 May 2015 • Adriana Kovashka, Kristen Grauman
We propose to discover shades of attribute meaning.