The objective of Audio-Visual Segmentation (AVS) is to locate sounding objects within visual scenes by accurately predicting pixelwise segmentation masks. In this paper, we present the following contributions: (i), we propose a scalable and annotation-free pipeline for generating artificial data for the AVS task. We leverage existing image segmentation and audio datasets to draw links between category labels, image-mask pairs, and audio samples, which allows us to easily compose (image, audio, mask) triplets for training AVS models; (ii), we introduce a novel Audio-Aware Transformer (AuTR) architecture that features an audio-aware query-based transformer decoder. This architecture enables the model to search for sounding objects with the guidance of audio signals, resulting in more accurate segmentation; (iii), we present extensive experiments conducted on both synthetic and real datasets, which demonstrate the effectiveness of training AVS models with synthetic data generated by our proposed pipeline. Additionally, our proposed AuTR architecture exhibits superior performance and strong generalization ability on public benchmarks. The project page is https://jinxiang-liu.github.io/anno-free-AVS/.
Large Language Models (LLMs) have showcased remarkable capabilities in natural language understanding in various domains. These models can usually behave well on daily dialog, or question answering scenarios, however, in areas that value precision, for example, in medical applications, they often exhibit unsatisfactory performance due to a lack of domain-specific knowledge. In this report, we introduce PMC-LLaMA, an open-source language model that is acquired by fine-tuning an open-source language model on a total of 4.8 million biomedical academic papers for further injecting medical knowledge, enhancing its capability in medical domain. Our preliminary evaluations are conducted on three biomedical QA datasets, including PubMedQA, MedMCQA, and USMLE, showing that the our model after finetuning, i.e., PMC-LLaMA, demonstrates better understanding of biomedical domain-specific concepts, thus achieving high performance on QA benchmarks. The model and codes, along with an online demo, are publicly available.
Segmentation is a core computer vision competency, with applications spanning a broad range of scientifically and economically valuable domains. To date, however, the prohibitive cost of annotation has limited the deployment of flexible segmentation models. In this work, we propose Zero-shot Unsupervised Transfer Instance Segmentation (ZUTIS), a framework that aims to meet this challenge. The key strengths of ZUTIS are: (i) no requirement for instance-level or pixel-level annotations; (ii) an ability of zero-shot transfer, i.e., no assumption on access to a target data distribution; (iii) a unified framework for semantic and instance segmentations with solid performance on both tasks compared to state-of-the-art unsupervised methods. While comparing to previous work, we show ZUTIS achieves a gain of 2.2 mask AP on COCO-20K and 14.5 mIoU on ImageNet-S with 919 categories for instance and semantic segmentations, respectively. The code is made publicly available.
Video Instance Segmentation(VIS) aims at segmenting and categorizing objects in videos from a closed set of training categories, lacking the generalization ability to handle novel categories in real-world videos. To address this limitation, we make the following three contributions. First, we introduce the novel task of Open-Vocabulary Video Instance Segmentation, which aims to simultaneously segment, track, and classify objects in videos from open-set categories, including novel categories unseen during training. Second, to benchmark Open-Vocabulary VIS, we collect a Large-Vocabulary Video Instance Segmentation dataset(LV-VIS), that contains well-annotated objects from 1,212 diverse categories, significantly surpassing the category size of existing datasets by more than one order of magnitude. Third, we propose an efficient Memory-Induced Vision-Language Transformer, MindVLT, to first achieve Open-Vocabulary VIS in an end-to-end manner with near real-time inference speed. Extensive experiments on LV-VIS and four existing VIS datasets demonstrate the strong zero-shot generalization ability of MindVLT on novel categories. We will release the dataset and code to facilitate future endeavors.
The objective of this paper is an automatic Audio Description (AD) model that ingests movies and outputs AD in text form. Generating high-quality movie AD is challenging due to the dependency of the descriptions on context, and the limited amount of training data available. In this work, we leverage the power of pretrained foundation models, such as GPT and CLIP, and only train a mapping network that bridges the two models for visually-conditioned text generation. In order to obtain high-quality AD, we make the following four contributions: (i) we incorporate context from the movie clip, AD from previous clips, as well as the subtitles; (ii) we address the lack of training data by pretraining on large-scale datasets, where visual or contextual information is unavailable, e.g. text-only AD without movies or visual captioning datasets without context; (iii) we improve on the currently available AD datasets, by removing label noise in the MAD dataset, and adding character naming information; and (iv) we obtain strong results on the movie AD task compared with previous methods.
Camera-only 3D detection provides an economical solution with a simple configuration for localizing objects in 3D space compared to LiDAR-based detection systems. However, a major challenge lies in precise depth estimation due to the lack of direct 3D measurements in the input. Many previous methods attempt to improve depth estimation through network designs, e.g., deformable layers and larger receptive fields. This work proposes an orthogonal direction, improving the camera-only 3D detection by introducing multi-agent collaborations. Our proposed collaborative camera-only 3D detection (CoCa3D) enables agents to share complementary information with each other through communication. Meanwhile, we optimize communication efficiency by selecting the most informative cues. The shared messages from multiple viewpoints disambiguate the single-agent estimated depth and complement the occluded and long-range regions in the single-agent view. We evaluate CoCa3D in one real-world dataset and two new simulation datasets. Results show that CoCa3D improves previous SOTA performances by 44.21% on DAIR-V2X, 30.60% on OPV2V+, 12.59% on CoPerception-UAVs+ for AP@70. Our preliminary results show a potential that with sufficient collaboration, the camera might overtake LiDAR in some practical scenarios. We released the dataset and code at https://siheng-chen.github.io/dataset/CoPerception+ and https://github.com/MediaBrain-SJTU/CoCa3D.
In this paper, we consider the problem of temporal action localization under low-shot (zero-shot & few-shot) scenario, with the goal of detecting and classifying the action instances from arbitrary categories within some untrimmed videos, even not seen at training time. We adopt a Transformer-based two-stage action localization architecture with class-agnostic action proposal, followed by open-vocabulary classification. We make the following contributions. First, to compensate image-text foundation models with temporal motions, we improve category-agnostic action proposal by explicitly aligning embeddings of optical flows, RGB and texts, which has largely been ignored in existing low-shot methods. Second, to improve open-vocabulary action classification, we construct classifiers with strong discriminative power, i.e., avoid lexical ambiguities. To be specific, we propose to prompt the pre-trained CLIP text encoder either with detailed action descriptions (acquired from large-scale language models), or visually-conditioned instance-specific prompt vectors. Third, we conduct thorough experiments and ablation studies on THUMOS14 and ActivityNet1.3, demonstrating the superior performance of our proposed model, outperforming existing state-of-the-art approaches by one significant margin.
Foundation models trained on large-scale dataset gain a recent surge in CV and NLP. In contrast, development in biomedical domain lags far behind due to data scarcity. To address this issue, we build and release PMC-OA, a biomedical dataset with 1.6M image-caption pairs collected from PubMedCentral's OpenAccess subset, which is 8 times larger than before. PMC-OA covers diverse modalities or diseases, with majority of the image-caption samples aligned at finer-grained level, i.e., subfigure and subcaption. While pretraining a CLIP-style model on PMC-OA, our model named PMC-CLIP achieves state-of-the-art results on various downstream tasks, including image-text retrieval on ROCO, MedMNIST image classification, Medical VQA, i.e. +8.1% R@10 on image-text retrieval, +3.9% accuracy on image classification.
Despite of the success of multi-modal foundation models pre-trained on large-scale data in natural language understanding and vision recognition, its counterpart in medical and clinical domains remains preliminary, due to the fine-grained recognition nature of the medical tasks with high demands on domain knowledge. Here, we propose a knowledge-enhanced vision-language pre-training approach for auto-diagnosis on chest X-ray images. The algorithm, named Knowledge-enhanced Auto Diagnosis~(KAD), first trains a knowledge encoder based on an existing medical knowledge graph, i.e., learning neural embeddings of the definitions and relationships between medical concepts and then leverages the pre-trained knowledge encoder to guide the visual representation learning with paired chest X-rays and radiology reports. We experimentally validate KAD's effectiveness on three external X-ray datasets. The zero-shot performance of KAD is not only comparable to that of the fully-supervised models but also, for the first time, superior to the average of three expert radiologists for three (out of five) pathologies with statistical significance. When the few-shot annotation is available, KAD also surpasses all existing approaches in finetuning settings, demonstrating the potential for application in different clinical scenarios.
In this paper, we consider the problem of disease diagnosis. Unlike the conventional learning paradigm that treats labels independently, we propose a knowledge-enhanced framework, that enables training visual representation with the guidance of medical domain knowledge. In particular, we make the following contributions: First, to explicitly incorporate experts' knowledge, we propose to learn a neural representation for the medical knowledge graph via contrastive learning, implicitly establishing relations between different medical concepts. Second, while training the visual encoder, we keep the parameters of the knowledge encoder frozen and propose to learn a set of prompt vectors for efficient adaptation. Third, we adopt a Transformer-based disease-query module for cross-model fusion, which naturally enables explainable diagnosis results via cross attention. To validate the effectiveness of our proposed framework, we conduct thorough experiments on three x-ray imaging datasets across different anatomy structures, showing our model is able to exploit the implicit relations between diseases/findings, thus is beneficial to the commonly encountered problem in the medical domain, namely, long-tailed and zero-shot recognition, which conventional methods either struggle or completely fail to realize.