An official website of the United States government
This paper proposes a modified method for training tool segmentation networks for endoscopic images by parsing training images into two disjoint sets: one for rectangular representations of endoscopic images and one for polar. Previous work [1], [2] demonstrated that certain endoscopic images may be better segmented by a U-Net network trained on the original rectangular representation of images alone, and others performed better with polar representations. This work extends that observation to the training images and seeks to intelligently decompose the aggregate training data into disjoint image sets — one ideal for training a network to segment original, rectangular endoscopic images and the other for training a polar segmentation network. The training set decomposition consists of three stages: (1) initial data split and models, (2) image reallocation and transition mechanisms with retraining, and (3) evaluation. In (2), two separate frameworks for parsing polar vs. rectangular training images were investigated, with three switching metrics utilized in both. Experiments comparatively evaluated the segmentation performance (via Sørenson Dice coefficient) of the in-group and out-of-group images between the set-decomposed models. Results are encouraging, showing improved aggregate in-group Dice scores as well as image sets trending towards convergence.
more »
« less
Experience with Using Synthetic Training Images for Wearable Cognitive Assistance
Wearable Cognitive Assistance (WCA) applications use computer vision models that require thousands of labeled training images. Capturing and labeling these images requires a substantial amount of work. By using synthetically generated images for training, we avoided this labor-intensive step. The performance of these models was comparable to that of models that were trained on real images.
more »
« less
- Award ID(s):
- 2106862
- PAR ID:
- 10409690
- Date Published:
- Journal Name:
- Proceedings of the AAAI Spring Symposium on AI Engineering, 2022
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Adversarial training is a popular defense strategy against attack threat models with bounded Lp norms. However, it often degrades the model performance on normal images and the defense does not generalize well to novel attacks. Given the success of deep generative models such as GANs and VAEs in characterizing the underlying manifold of images, we investigate whether or not the aforementioned problems can be remedied by exploiting the underlying manifold information. To this end, we construct an "On-Manifold ImageNet" (OM-ImageNet) dataset by projecting the ImageNet samples onto the manifold learned by StyleGSN. For this dataset, the underlying manifold information is exact. Using OM-ImageNet, we first show that adversarial training in the latent space of images improves both standard accuracy and robustness to on-manifold attacks. However, since no out-of-manifold perturbations are realized, the defense can be broken by Lp adversarial attacks. We further propose Dual Manifold Adversarial Training (DMAT) where adversarial perturbations in both latent and image spaces are used in robustifying the model. Our DMAT improves performance on normal images, and achieves comparable robustness to the standard adversarial training against Lp attacks. In addition, we observe that models defended by DMAT achieve improved robustness against novel attacks which manipulate images by global color shifts or various types of image filtering. Interestingly, similar improvements are also achieved when the defended models are tested on out-of-manifold natural images. These results demonstrate the potential benefits of using manifold information in enhancing robustness of deep learning models against various types of novel adversarial attacks.more » « less
-
Modern vision models typically rely on fine-tuning general-purpose models pre-trained on large, static datasets. These general-purpose models only capture the knowledge within their pre-training datasets, which are tiny, out-of-date snapshots of the Internet -- where billions of images are uploaded each day. We suggest an alternate approach: rather than hoping our static datasets transfer to our desired tasks after large-scale pre-training, we propose dynamically utilizing the Internet to quickly train a small-scale model that does extremely well on the task at hand. Our approach, called Internet Explorer, explores the web in a self-supervised manner to progressively find relevant examples that improve performance on a desired target dataset. It cycles between searching for images on the Internet with text queries, self-supervised training on downloaded images, determining which images were useful, and prioritizing what to search for next. We evaluate Internet Explorer across several datasets and show that it outperforms or matches CLIP oracle performance by using just a single GPU desktop to actively query the Internet for 30--40 hours.more » « less
-
Diffusion models excel at generating photo-realistic images but come with significant computational costs in both training and sampling. While various techniques address these computational challenges, a less-explored issue is designing an efficient and adaptable network backbone for iterative refinement. Current options like U-Net and Vision Transformer often rely on resource-intensive deep networks and lack the flexibility needed for generating images at variable resolutions or with a smaller network than used in training. This study introduces LEGO bricks, which seamlessly integrate Local-feature Enrichment and Global-content Orchestration. These bricks can be stacked to create a test-time reconfigurable diffusion backbone, allowing selective skipping of bricks to reduce sampling costs and generate higher-resolution images than the training data. LEGO bricks enrich local regions with an MLP and transform them using a Transformer block while maintaining a consistent full-resolution image across all bricks. Experimental results demonstrate that LEGO bricks enhance training efficiency, expedite convergence, and facilitate variable-resolution image generation while maintaining strong generative performance. Moreover, LEGO significantly reduces sampling time compared to other methods, establishing it as a valuable enhancement for diffusion models.more » « less
-
Today, creators of data-hungry deep neural networks (DNNs) scour the Internet for training fodder, leaving users with little control over or knowledge of when their data, and in particular their images, are used to train models. To empower users to counteract unwanted use of their images, we design, implement and evaluate a practical system that enables users to detect if their data was used to train a DNN model for image classification. We show how users can create special images we call isotopes, which introduce ``spurious features'' into DNNs during training. With only query access to a model and no knowledge of the model-training process, nor control of the data labels, a user can apply statistical hypothesis testing to detect if the model learned these spurious features by training on the user's images. Isotopes can be viewed as an application of a particular type of data poisoning. In contrast to backdoors and other poisoning attacks, our purpose is not to cause misclassification but rather to create tell-tale changes in confidence scores output by the model that reveal the presence of isotopes in the training data. Isotopes thus turn DNNs' vulnerability to memorization and spurious correlations into a tool for data provenance. Our results confirm efficacy in multiple image classification settings, detecting and distinguishing between hundreds of isotopes with high accuracy. We further show that our system works on public ML-as-a-service platforms and larger models such as ImageNet, can use physical objects in images instead of digital marks, and remains robust against several adaptive countermeasures.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Conference Paper:
- The DOI is not currently available.
