Scene analysis is essential for enabling autonomous systems, such as mobile robots, to operate in real-world environments. However, obtaining a comprehensive understanding of the scene requires solving multiple tasks, such as panoptic segmentation, instance orientation estimation, and scene classification. Solving these tasks given limited computing and battery capabilities on mobile platforms is challenging. To address this challenge, we introduce an efficient multi-task scene analysis approach, called EMSAFormer, that uses an RGB-D Transformer-based encoder to simultaneously perform the aforementioned tasks. Our approach builds upon the previously published EMSANet. However, we show that the dual CNN-based encoder of EMSANet can be replaced with a single Transformer-based encoder. To achieve this, we investigate how information from both RGB and depth data can be effectively incorporated in a single encoder. To accelerate inference on robotic hardware, we provide a custom NVIDIA TensorRT extension enabling highly optimization for our EMSAFormer approach. Through extensive experiments on the commonly used indoor datasets NYUv2, SUNRGB-D, and ScanNet, we show that our approach achieves state-of-the-art performance while still enabling inference with up to 39.1 FPS on an NVIDIA Jetson AGX Orin 32 GB.
Exploring robust and efficient association methods has always been an important issue in multiple-object tracking (MOT). Although existing tracking methods have achieved impressive performance, congestion and frequent occlusions still pose challenging problems in multi-object tracking. We reveal that performing sparse decomposition on dense scenes is a crucial step to enhance the performance of associating occluded targets. To this end, we propose a pseudo-depth estimation method for obtaining the relative depth of targets from 2D images. Secondly, we design a depth cascading matching (DCM) algorithm, which can use the obtained depth information to convert a dense target set into multiple sparse target subsets and perform data association on these sparse target subsets in order from near to far. By integrating the pseudo-depth method and the DCM strategy into the data association process, we propose a new tracker, called SparseTrack. SparseTrack provides a new perspective for solving the challenging crowded scene MOT problem. Only using IoU matching, SparseTrack achieves comparable performance with the state-of-the-art (SOTA) methods on the MOT17 and MOT20 benchmarks. Code and models are publicly available at \url{https://github.com/hustvl/SparseTrack}.
Tabular data is often hidden in text, particularly in medical diagnostic reports. Traditional machine learning (ML) models designed to work with tabular data, cannot effectively process information in such form. On the other hand, large language models (LLMs) which excel at textual tasks, are probably not the best tool for modeling tabular data. Therefore, we propose a novel, simple, and effective methodology for extracting structured tabular data from textual medical reports, called TEMED-LLM. Drawing upon the reasoning capabilities of LLMs, TEMED-LLM goes beyond traditional extraction techniques, accurately inferring tabular features, even when their names are not explicitly mentioned in the text. This is achieved by combining domain-specific reasoning guidelines with a proposed data validation and reasoning correction feedback loop. By applying interpretable ML models such as decision trees and logistic regression over the extracted and validated data, we obtain end-to-end interpretable predictions. We demonstrate that our approach significantly outperforms state-of-the-art text classification models in medical diagnostics. Given its predictive performance, simplicity, and interpretability, TEMED-LLM underscores the potential of leveraging LLMs to improve the performance and trustworthiness of ML models in medical applications.
Reenacting facial images is an important task that can find numerous applications. We proposed IFaceUV, a fully differentiable pipeline that properly combines 2D and 3D information to conduct the facial reenactment task. The three-dimensional morphable face models (3DMMs) and corresponding UV maps are utilized to intuitively control facial motions and textures, respectively. Two-dimensional techniques based on 2D image warping is further required to compensate for missing components of the 3DMMs such as backgrounds, ear, hair and etc. In our pipeline, we first extract 3DMM parameters and corresponding UV maps from source and target images. Then, initial UV maps are refined by the UV map refinement network and it is rendered to the image with the motion manipulated 3DMM parameters. In parallel, we warp the source image according to the 2D flow field obtained from the 2D warping network. Rendered and warped images are combined in the final editing network to generate the final reenactment image. Additionally, we tested our model for the audio-driven facial reenactment task. Extensive qualitative and quantitative experiments illustrate the remarkable performance of our method compared to other state-of-the-art methods.
Sequence labeling tasks require the computation of sentence representations for each word within a given sentence. With the rise of advanced pretrained language models; one common approach involves incorporating a BiLSTM layer to enhance the sequence structure information at the output level. Nevertheless, it has been empirically demonstrated (P.-H. Li, 2020) that BiLSTM's potential for generating sentence representations for sequence labeling tasks is constrained, primarily due to the integration of fragments from past and future sentence representations to form a complete sentence representation. In this study, we observed that the entire sentence representation, found in both the first and last cells of BiLSTM, can supplement each cell's sentence representation. Accordingly, we devised a global context mechanism to integrate entire future and past sentence representations into each cell's sentence representation within BiLSTM, leading to a significant improvement in both F1 score and accuracy. By embedding the BERT model within BiLSTM as a demonstration, and conducting exhaustive experiments on nine datasets for sequence labeling tasks, including named entity recognition (NER), part of speech (POS) tagging and End-to-End Aspect-Based sentiment analysis (E2E-ABSA). We noted significant improvements in F1 scores and accuracy across all examined datasets.
The posterior collapse phenomenon in variational autoencoders (VAEs), where the variational posterior distribution closely matches the prior distribution, can hinder the quality of the learned latent variables. As a consequence of posterior collapse, the latent variables extracted by the encoder in VAEs preserve less information from the input data and thus fail to produce meaningful representations as input to the reconstruction process in the decoder. While this phenomenon has been an actively addressed topic related to VAEs performance, the theory for posterior collapse remains underdeveloped, especially beyond the standard VAEs. In this work, we advance the theoretical understanding of posterior collapse to two important and prevalent yet less studied classes of VAEs: conditional VAEs and hierarchical VAEs. Specifically, via a non-trivial theoretical analysis of linear conditional VAEs and hierarchical VAEs with two levels of latent, we prove that the cause of posterior collapses in these models includes the correlation between the input and output of the conditional VAEs and the effect of learnable encoder variance in the hierarchical VAEs. We empirically validate our theoretical findings for linear conditional and hierarchical VAEs and demonstrate that these results are also predictive for non-linear cases.
Existing offboard 3D detectors always follow a modular pipeline design to take advantage of unlimited sequential point clouds. We have found that the full potential of offboard 3D detectors is not explored mainly due to two reasons: (1) the onboard multi-object tracker cannot generate sufficient complete object trajectories, and (2) the motion state of objects poses an inevitable challenge for the object-centric refining stage in leveraging the long-term temporal context representation. To tackle these problems, we propose a novel paradigm of offboard 3D object detection, named DetZero. Concretely, an offline tracker coupled with a multi-frame detector is proposed to focus on the completeness of generated object tracks. An attention-mechanism refining module is proposed to strengthen contextual information interaction across long-term sequential point clouds for object refining with decomposed regression methods. Extensive experiments on Waymo Open Dataset show our DetZero outperforms all state-of-the-art onboard and offboard 3D detection methods. Notably, DetZero ranks 1st place on Waymo 3D object detection leaderboard with 85.15 mAPH (L2) detection performance. Further experiments validate the application of taking the place of human labels with such high-quality results. Our empirical study leads to rethinking conventions and interesting findings that can guide future research on offboard 3D object detection.
Large Language Models (LLMs) store an extensive amount of factual knowledge obtained from vast collections of text. To effectively utilize these models for downstream tasks, it is crucial to have reliable methods for measuring their knowledge. However, existing approaches for knowledge measurement have certain limitations, and despite recent efforts, they fail to provide accurate measurements and the necessary insights for modifying the knowledge within LLMs. In this work, we employ information theory-based measurements to provide a framework estimating the factual knowledge contained within large language models. More specifically, we measure knowledge by analyzing the LLM's prediction probability distribution before and after instilling the target knowledge, employing metrics such as entropy and KL-divergence. Introducing our metrics, we first assess their accuracy in comparison to previous ranking-based methods, surpassing them by over $35\%$ in a synthetic experiment. Then, we explore two prominent methods of knowledge instillation, discovering that LLMs exhibit limitations in capturing new knowledge under specific circumstances for one of these methods. Lastly, we demonstrate the applicability of our methods in extracting unlearned and mislearned facts in LLMs through their application to in-context learning. We make code and data for all methods and experiments in this paper publicly available.
Side information is being used extensively to improve the effectiveness of sequential recommendation models. It is said to help capture the transition patterns among items. Most previous work on sequential recommendation that uses side information models item IDs and side information separately. This can only model part of relations between items and their side information. Moreover, in real-world systems, not all values of item feature fields are available. This hurts the performance of models that rely on side information. Existing methods tend to neglect the context of missing item feature fields, and fill them with generic or special values, e.g., unknown, which might lead to sub-optimal performance. To address the limitation of sequential recommenders with side information, we define a way to fuse side information and alleviate the problem of missing side information by proposing a unified task, namely the missing information imputation (MII), which randomly masks some feature fields in a given sequence of items, including item IDs, and then forces a predictive model to recover them. By considering the next item as a missing feature field, sequential recommendation can be formulated as a special case of MII. We propose a sequential recommendation model, called missing information imputation recommender (MIIR), that builds on the idea of MII and simultaneously imputes missing item feature values and predicts the next item. We devise a dense fusion self-attention (DFSA) for MIIR to capture all pairwise relations between items and their side information. Empirical studies on three benchmark datasets demonstrate that MIIR, supervised by MII, achieves a significantly better sequential recommendation performance than state-of-the-art baselines.
State-of-the-art approaches to speaker anonymization typically employ some form of perturbation function to conceal speaker information contained within an x-vector embedding, then resynthesize utterances in the voice of a new pseudo-speaker using a vocoder. Strategies to improve the x-vector anonymization function have attracted considerable research effort, whereas vocoder impacts are generally neglected. In this paper, we show that the impact of the vocoder is substantial and sometimes dominant. The vocoder drift, namely the difference between the x-vector vocoder input and that which can be extracted subsequently from the output, is learnable and can hence be reversed by an attacker; anonymization can be undone and the level of privacy protection provided by such approaches might be weaker than previously thought. The findings call into question the focus upon x-vector anonymization, prompting the need for greater attention to vocoder impacts and stronger attack models alike.