A significant challenge in the field of object detection lies in the system's performance under non-ideal imaging conditions, such as rain, fog, low illumination, or raw Bayer images that lack ISP processing. Our study introduces "Feature Corrective Transfer Learning", a novel approach that leverages transfer learning and a bespoke loss function to facilitate the end-to-end detection of objects in these challenging scenarios without the need to convert non-ideal images into their RGB counterparts. In our methodology, we initially train a comprehensive model on a pristine RGB image dataset. Subsequently, non-ideal images are processed by comparing their feature maps against those from the initial ideal RGB model. This comparison employs the Extended Area Novel Structural Discrepancy Loss (EANSDL), a novel loss function designed to quantify similarities and integrate them into the detection loss. This approach refines the model's ability to perform object detection across varying conditions through direct feature map correction, encapsulating the essence of Feature Corrective Transfer Learning. Experimental validation on variants of the KITTI dataset demonstrates a significant improvement in mean Average Precision (mAP), resulting in a 3.8-8.1% relative enhancement in detection under non-ideal conditions compared to the baseline model, and a less marginal performance difference within 1.3% of the mAP@[0.5:0.95] achieved under ideal conditions by the standard Faster RCNN algorithm.
相關內容
Expressive speech-to-speech translation (S2ST) is a key research topic in seamless communication, which focuses on the preservation of semantics and speaker vocal style in translated speech. Early works synthesized speaker style aligned speech in order to directly learn the mapping from speech to target speech spectrogram. Without reliance on style aligned data, recent studies leverage the advances of language modeling (LM) and build cascaded LMs on semantic and acoustic tokens. This work proposes SeamlessExpressiveLM, a single speech language model for expressive S2ST. We decompose the complex source-to-target speech mapping into intermediate generation steps with chain-of-thought prompting. The model is first guided to translate target semantic content and then transfer the speaker style to multi-stream acoustic units. Evaluated on Spanish-to-English and Hungarian-to-English translations, SeamlessExpressiveLM outperforms cascaded LMs in both semantic quality and style transfer, meanwhile achieving better parameter efficiency.
Human pose estimation is a key task in computer vision with various applications such as activity recognition and interactive systems. However, the lack of consistency in the annotated skeletons across different datasets poses challenges in developing universally applicable models. To address this challenge, we propose a novel approach integrating multi-teacher knowledge distillation with a unified skeleton representation. Our networks are jointly trained on the COCO and MPII datasets, containing 17 and 16 keypoints, respectively. We demonstrate enhanced adaptability by predicting an extended set of 21 keypoints, 4 (COCO) and 5 (MPII) more than original annotations, improving cross-dataset generalization. Our joint models achieved an average accuracy of 70.89 and 76.40, compared to 53.79 and 55.78 when trained on a single dataset and evaluated on both. Moreover, we also evaluate all 21 predicted points by our two models by reporting an AP of 66.84 and 72.75 on the Halpe dataset. This highlights the potential of our technique to address one of the most pressing challenges in pose estimation research and application - the inconsistency in skeletal annotations.
Learning holistic computational representations in physical, chemical or biological systems requires the ability to process information from different distributions and modalities within the same model. Thus, the demand for multimodal machine learning models has sharply risen for modalities that go beyond vision and language, such as sequences, graphs, time series, or tabular data. While there are many available multimodal fusion and alignment approaches, most of them require end-to-end training, scale quadratically with the number of modalities, cannot handle cases of high modality imbalance in the training set, or are highly topology-specific, making them too restrictive for many biomedical learning tasks. This paper presents Multimodal Lego (MM-Lego), a modular and general-purpose fusion and model merging framework to turn any set of encoders into a competitive multimodal model with no or minimal fine-tuning. We achieve this by introducing a wrapper for unimodal encoders that enforces lightweight dimensionality assumptions between modalities and harmonises their representations by learning features in the frequency domain to enable model merging with little signal interference. We show that MM-Lego 1) can be used as a model merging method which achieves competitive performance with end-to-end fusion models without any fine-tuning, 2) can operate on any unimodal encoder, and 3) is a model fusion method that, with minimal fine-tuning, achieves state-of-the-art results on six benchmarked multimodal biomedical tasks.
Open-vocabulary object detection (OVD) requires solid modeling of the region-semantic relationship, which could be learned from massive region-text pairs. However, such data is limited in practice due to significant annotation costs. In this work, we propose RTGen to generate scalable open-vocabulary region-text pairs and demonstrate its capability to boost the performance of open-vocabulary object detection. RTGen includes both text-to-region and region-to-text generation processes on scalable image-caption data. The text-to-region generation is powered by image inpainting, directed by our proposed scene-aware inpainting guider for overall layout harmony. For region-to-text generation, we perform multiple region-level image captioning with various prompts and select the best matching text according to CLIP similarity. To facilitate detection training on region-text pairs, we also introduce a localization-aware region-text contrastive loss that learns object proposals tailored with different localization qualities. Extensive experiments demonstrate that our RTGen can serve as a scalable, semantically rich, and effective source for open-vocabulary object detection and continue to improve the model performance when more data is utilized, delivering superior performance compared to the existing state-of-the-art methods.
While dynamic graph neural networks have shown promise in various applications, explaining their predictions on continuous-time dynamic graphs (CTDGs) is difficult. This paper investigates a new research task: self-interpretable GNNs for CTDGs. We aim to predict future links within the dynamic graph while simultaneously providing causal explanations for these predictions. There are two key challenges: (1) capturing the underlying structural and temporal information that remains consistent across both independent and identically distributed (IID) and out-of-distribution (OOD) data, and (2) efficiently generating high-quality link prediction results and explanations. To tackle these challenges, we propose a novel causal inference model, namely the Independent and Confounded Causal Model (ICCM). ICCM is then integrated into a deep learning architecture that considers both effectiveness and efficiency. Extensive experiments demonstrate that our proposed model significantly outperforms existing methods across link prediction accuracy, explanation quality, and robustness to shortcut features. Our code and datasets are anonymously released at //github.com/2024SIG/SIG.
We derive a new parallel-in-time approach for solving large-scale optimization problems constrained by time-dependent partial differential equations arising from fluid dynamics. The solver involves the use of a block circulant approximation of the original matrices, enabling parallelization-in-time via the use of fast Fourier transforms, and we devise bespoke matrix approximations which may be applied within this framework. These make use of permutations, saddle-point approximations, commutator arguments, as well as inner solvers such as the Uzawa method, Chebyshev semi-iteration, and multigrid. Theoretical results underpin our strategy of applying a block circulant strategy, and numerical experiments demonstrate the effectiveness and robustness of our approach on Stokes and Oseen problems. Noteably, satisfying results for the strong and weak scaling of our methods are provided within a fully parallel architecture.
The biased net paradigm was the first general and empirically tractable scheme for parameterizing complex patterns of dependence in networks, expressing deviations from uniform random graph structure in terms of latent ``bias events,'' whose realizations enhance reciprocity, transitivity, or other structural features. Subsequent developments have introduced local specifications of biased nets, which reduce the need for approximations required in early specifications based on tracing processes. Here, we show that while one such specification leads to inconsistencies, a closely related Markovian specification both evades these difficulties and can be extended to incorporate new types of effects. We introduce the notion of inhibitory bias events, with satiation as an example, which are useful for avoiding degeneracies that can arise from closure bias terms. Although our approach does not lead to a computable likelihood, we provide a strategy for approximate Bayesian inference using random forest prevision. We demonstrate our approach on a network of friendship ties among college students, recapitulating a relationship between the sibling bias and tie strength posited in earlier work by Fararo.
Due to the technical complexity and social impact, automated vehicle (AV) development challenges the current state of automotive engineering practice. Research shows that it is important to consider human factors (HF) knowledge when developing AVs to make them safe and accepted. This study explores the current practices and challenges of the automotive industries for incorporating HF requirements during agile AV development. We interviewed ten industry professionals from several Swedish automotive companies, including HF experts and AV engineers. Based on our qualitative analysis of the semi-structured interviews, a number of current approaches for communicating and incorporating HF knowledge into agile AV development and associated challenges are discussed. Our findings may help to focus future research on issues that are critical to effectively incorporate HF knowledge into agile AV development.
Face recognition technology has advanced significantly in recent years due largely to the availability of large and increasingly complex training datasets for use in deep learning models. These datasets, however, typically comprise images scraped from news sites or social media platforms and, therefore, have limited utility in more advanced security, forensics, and military applications. These applications require lower resolution, longer ranges, and elevated viewpoints. To meet these critical needs, we collected and curated the first and second subsets of a large multi-modal biometric dataset designed for use in the research and development (R&D) of biometric recognition technologies under extremely challenging conditions. Thus far, the dataset includes more than 350,000 still images and over 1,300 hours of video footage of approximately 1,000 subjects. To collect this data, we used Nikon DSLR cameras, a variety of commercial surveillance cameras, specialized long-rage R&D cameras, and Group 1 and Group 2 UAV platforms. The goal is to support the development of algorithms capable of accurately recognizing people at ranges up to 1,000 m and from high angles of elevation. These advances will include improvements to the state of the art in face recognition and will support new research in the area of whole-body recognition using methods based on gait and anthropometry. This paper describes methods used to collect and curate the dataset, and the dataset's characteristics at the current stage.
Existing recommender systems extract the user preference based on learning the correlation in data, such as behavioral correlation in collaborative filtering, feature-feature, or feature-behavior correlation in click-through rate prediction. However, regretfully, the real world is driven by causality rather than correlation, and correlation does not imply causation. For example, the recommender systems can recommend a battery charger to a user after buying a phone, in which the latter can serve as the cause of the former, and such a causal relation cannot be reversed. Recently, to address it, researchers in recommender systems have begun to utilize causal inference to extract causality, enhancing the recommender system. In this survey, we comprehensively review the literature on causal inference-based recommendation. At first, we present the fundamental concepts of both recommendation and causal inference as the basis of later content. We raise the typical issues that the non-causality recommendation is faced. Afterward, we comprehensively review the existing work of causal inference-based recommendation, based on a taxonomy of what kind of problem causal inference addresses. Last, we discuss the open problems in this important research area, along with interesting future works.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191