Well-formed context aware image captions and tags in enterprise content such as marketing material are critical to ensure their brand presence and content recall. Manual creation and updates to ensure the same is non trivial given the scale and the tedium towards this task. We propose a new unified Vision-Language (VL) model based on the One For All (OFA) model, with a focus on context-assisted image captioning where the caption is generated based on both the image and its context. Our approach aims to overcome the context-independent (image and text are treated independently) nature of the existing approaches. We exploit context by pretraining our model with datasets of three tasks: news image captioning where the news article is the context, contextual visual entailment, and keyword extraction from the context. The second pretraining task is a new VL task, and we construct and release two datasets for the task with 1.1M and 2.2K data instances. Our system achieves state-of-the-art results with an improvement of up to 8.34 CIDEr score on the benchmark news image captioning datasets. To the best of our knowledge, ours is the first effort at incorporating contextual information in pretraining the models for the VL tasks.
It has long been an ill-posed problem to predict absolute depth maps from single images in real (unseen) indoor scenes. We observe that it is essentially due to not only the scale-ambiguous problem but also the focal-ambiguous problem that decreases the generalization ability of monocular depth estimation. That is, images may be captured by cameras of different focal lengths in scenes of different scales. In this paper, we develop a focal-and-scale depth estimation model to well learn absolute depth maps from single images in unseen indoor scenes. First, a relative depth estimation network is adopted to learn relative depths from single images with diverse scales/semantics. Second, multi-scale features are generated by mapping a single focal length value to focal length features and concatenating them with intermediate features of different scales in relative depth estimation. Finally, relative depths and multi-scale features are jointly fed into an absolute depth estimation network. In addition, a new pipeline is developed to augment the diversity of focal lengths of public datasets, which are often captured with cameras of the same or similar focal lengths. Our model is trained on augmented NYUDv2 and tested on three unseen datasets. Our model considerably improves the generalization ability of depth estimation by 41%/13% (RMSE) with/without data augmentation compared with five recent SOTAs and well alleviates the deformation problem in 3D reconstruction. Notably, our model well maintains the accuracy of depth estimation on original NYUDv2.
We examine whether symmetry can be used to defend tree-based ensemble classifiers such as gradient-boosting decision trees (GBDTs) against adversarial perturbation attacks. The idea is based on a recent symmetry defense for convolutional neural network classifiers (CNNs) that utilizes CNNs' lack of invariance with respect to symmetries. CNNs lack invariance because they can classify a symmetric sample, such as a horizontally flipped image, differently from the original sample. CNNs' lack of invariance also means that CNNs can classify symmetric adversarial samples differently from the incorrect classification of adversarial samples. Using CNNs' lack of invariance, the recent CNN symmetry defense has shown that the classification of symmetric adversarial samples reverts to the correct sample classification. In order to apply the same symmetry defense to GBDTs, we examine GBDT invariance and are the first to show that GBDTs also lack invariance with respect to symmetries. We apply and evaluate the GBDT symmetry defense for nine datasets against six perturbation attacks with a threat model that ranges from zero-knowledge to perfect-knowledge adversaries. Using the feature inversion symmetry against zero-knowledge adversaries, we achieve up to 100% accuracy on adversarial samples even when default and robust classifiers have 0% accuracy. Using the feature inversion and horizontal flip symmetries against perfect-knowledge adversaries, we achieve up to over 95% accuracy on adversarial samples for the GBDT classifier of the F-MNIST dataset even when default and robust classifiers have 0% accuracy.
Precise identification of multiple cell classes in high-resolution Giga-pixel whole slide imaging (WSI) is critical for various clinical scenarios. Building an AI model for this purpose typically requires pixel-level annotations, which are often unscalable and must be done by skilled domain experts (e.g., pathologists). However, these annotations can be prone to errors, especially when distinguishing between intricate cell types (e.g., podocytes and mesangial cells) using only visual inspection. Interestingly, a recent study showed that lay annotators, when using extra immunofluorescence (IF) images for reference (referred to as molecular-empowered learning), can sometimes outperform domain experts in labeling. Despite this, the resource-intensive task of manual delineation remains a necessity during the annotation process. In this paper, we explore the potential of bypassing pixel-level delineation by employing the recent segment anything model (SAM) on weak box annotation in a zero-shot learning approach. Specifically, we harness SAM's ability to produce pixel-level annotations from box annotations and utilize these SAM-generated labels to train a segmentation model. Our findings show that the proposed SAM-assisted molecular-empowered learning (SAM-L) can diminish the labeling efforts for lay annotators by only requiring weak box annotations. This is achieved without compromising annotation accuracy or the performance of the deep learning-based segmentation. This research represents a significant advancement in democratizing the annotation process for training pathological image segmentation, relying solely on non-expert annotators.
With the development of the multi-media internet, visual characteristics have become an important factor affecting user interests. Thus, incorporating visual features is a promising direction for further performance improvements in click-through rate (CTR) prediction. However, we found that simply injecting the image embeddings trained with established pre-training methods only has marginal improvements. We attribute the failure to two reasons: First, The pre-training methods are designed for well-defined computer vision tasks concentrating on semantic features, and they cannot learn personalized interest in recommendations. Secondly, pre-trained image embeddings only containing semantic information have little information gain, considering we already have semantic features such as categories and item titles as inputs in the CTR prediction task. We argue that a pre-training method tailored for recommendation is necessary for further improvements. To this end, we propose a recommendation-aware image pre-training method that can learn visual features from user click histories. Specifically, we propose a user interest reconstruction module to mine visual features related to user interests from behavior histories. We further propose a contrastive training method to avoid collapsing of embedding vectors. We conduct extensive experiments to verify that our method can learn users' visual interests, and our method achieves $0.46\%$ improvement in offline AUC and $0.88\%$ improvement in Taobao online GMV with p-value$<0.01$.
Purpose: Echo-planar imaging (EPI) with blip-up/down acquisition (BUDA) can provide high-quality images with minimal distortions by using two readout trains with opposing phase-encoding gradients. Because of the need for two separate acquisitions, BUDA doubles the scan time and degrades the temporal resolution when compared to single-shot EPI, presenting a major challenge for many applications, particularly functional MRI (fMRI). This study aims at overcoming this challenge by developing an echo-shifted EPI BUDA (esEPI-BUDA) technique to acquire both blip-up and blip-down datasets in a single shot. Methods: A three-dimensional (3D) esEPI-BUDA pulse sequence was designed by using an echo-shifting strategy to produce two EPI readout trains. These readout trains produced a pair of k-space datasets whose k-space trajectories were interleaved with opposite phase-encoding gradient directions. The two k-space datasets were separately reconstructed using a 3D SENSE algorithm, from which time-resolved B0-field maps were derived using TOPUP in FSL and then input into a forward model of joint parallel imaging reconstruction to correct for geometric distortion. In addition, Hankel structured low-rank constraint was incorporated into the reconstruction framework to improve image quality by mitigating the phase errors between the two interleaved k-space datasets. Results: The 3D esEPI-BUDA technique was demonstrated in a phantom and an fMRI study on healthy human subjects. Geometric distortions were effectively corrected in both phantom and human brain images. In the fMRI study, the visual activation volumes and their BOLD responses were comparable to those from conventional 3D echo-planar images. Conclusion: The improved imaging efficiency and dynamic distortion correction capability afforded by 3D esEPI-BUDA are expected to benefit many EPI applications.
Denoising diffusion probabilistic models that were initially proposed for realistic image generation have recently shown success in various perception tasks (e.g., object detection and image segmentation) and are increasingly gaining attention in computer vision. However, extending such models to multi-frame human pose estimation is non-trivial due to the presence of the additional temporal dimension in videos. More importantly, learning representations that focus on keypoint regions is crucial for accurate localization of human joints. Nevertheless, the adaptation of the diffusion-based methods remains unclear on how to achieve such objective. In this paper, we present DiffPose, a novel diffusion architecture that formulates video-based human pose estimation as a conditional heatmap generation problem. First, to better leverage temporal information, we propose SpatioTemporal Representation Learner which aggregates visual evidences across frames and uses the resulting features in each denoising step as a condition. In addition, we present a mechanism called Lookup-based MultiScale Feature Interaction that determines the correlations between local joints and global contexts across multiple scales. This mechanism generates delicate representations that focus on keypoint regions. Altogether, by extending diffusion models, we show two unique characteristics from DiffPose on pose estimation task: (i) the ability to combine multiple sets of pose estimates to improve prediction accuracy, particularly for challenging joints, and (ii) the ability to adjust the number of iterative steps for feature refinement without retraining the model. DiffPose sets new state-of-the-art results on three benchmarks: PoseTrack2017, PoseTrack2018, and PoseTrack21.
Text-only Image Captioning (TIC) is an approach that aims to construct a model solely based on text that can accurately describe images. Recently, diffusion models have demonstrated remarkable capabilities in generating high-quality images that are semantically coherent with given texts. This presents an opportunity to generate synthetic training images for TIC. However, we have identified a challenge that the images generated from simple descriptions typically exhibit a single perspective with one or limited contexts, which is not aligned with the complexity of real-world scenes in the image domain. In this paper, we propose a novel framework that addresses this issue by introducing multi-context data generation. Starting with an initial text corpus, our framework employs a large language model to select multiple sentences that describe the same scene from various perspectives. These sentences are then summarized into a single sentence with multiple contexts. We generate simple images using the straightforward sentences and complex images using the summarized sentences through diffusion models. Finally, we train the model exclusively using the synthetic image-text pairs obtained from this process. Experimental results demonstrate that our proposed framework effectively tackles the central challenge we have identified, achieving the state-of-the-art performance on popular datasets such as MSCOCO, Flickr30k, and SS1M.
In this paper, we propose a table and image generation task to verify how the knowledge about entities acquired from natural language is retained in Vision & Language (V & L) models. This task consists of two parts: the first is to generate a table containing knowledge about an entity and its related image, and the second is to generate an image from an entity with a caption and a table containing related knowledge of the entity. In both tasks, the model must know the entities used to perform the generation properly. We created the Wikipedia Table and Image Generation (WikiTIG) dataset from about 200,000 infoboxes in English Wikipedia articles to perform the proposed tasks. We evaluated the performance on the tasks with respect to the above research question using the V & L model OFA, which has achieved state-of-the-art results in multiple tasks. Experimental results show that OFA forgets part of its entity knowledge by pre-training as a complement to improve the performance of image related tasks.
Convolutional neural networks (CNNs) often suffer from poor performance when tested on target data that differs from the training (source) data distribution, particularly in medical imaging applications where variations in imaging protocols across different clinical sites and scanners lead to different imaging appearances. However, re-accessing source training data for unsupervised domain adaptation or labeling additional test data for model fine-tuning can be difficult due to privacy issues and high labeling costs, respectively. To solve this problem, we propose a novel atlas-guided test-time adaptation (TTA) method for robust 3D medical image segmentation, called AdaAtlas. AdaAtlas only takes one single unlabeled test sample as input and adapts the segmentation network by minimizing an atlas-based loss. Specifically, the network is adapted so that its prediction after registration is aligned with the learned atlas in the atlas space, which helps to reduce anatomical segmentation errors at test time. In addition, different from most existing TTA methods which restrict the adaptation to batch normalization blocks in the segmentation network only, we further exploit the use of channel and spatial attention blocks for improved adaptability at test time. Extensive experiments on multiple datasets from different sites show that AdaAtlas with attention blocks adapted (AdaAtlas-Attention) achieves superior performance improvements, greatly outperforming other competitive TTA methods.