Hyperspectral images have significant applications in various domains, since they register numerous semantic and spatial information in the spectral band with spatial variability of spectral signatures. Two critical challenges in identifying pixels of the hyperspectral image are respectively representing the correlated information among the local and global, as well as the abundant parameters of the model. To tackle this challenge, we propose a Multi-Scale U-shape Multi-Layer Perceptron (MUMLP) a model consisting of the designed MSC (Multi-Scale Channel) block and the UMLP (U-shape Multi-Layer Perceptron) structure. MSC transforms the channel dimension and mixes spectral band feature to embed the deep-level representation adequately. UMLP is designed by the encoder-decoder structure with multi-layer perceptron layers, which is capable of compressing the large-scale parameters. Extensive experiments are conducted to demonstrate our model can outperform state-of-the-art methods across-the-board on three wide-adopted public datasets, namely Pavia University, Houston 2013 and Houston 2018
Existing neural models are demonstrated to struggle with compositional generalization (CG), i.e., the ability to systematically generalize to unseen compositions of seen components. A key reason for failure on CG is that the syntactic and semantic representations of sequences in both the uppermost layer of the encoder and decoder are entangled. However, previous work concentrates on separating the learning of syntax and semantics instead of exploring the reasons behind the representation entanglement (RE) problem to solve it. We explain why it exists by analyzing the representation evolving mechanism from the bottom to the top of the Transformer layers. We find that the ``shallow'' residual connections within each layer fail to fuse previous layers' information effectively, leading to information forgetting between layers and further the RE problems. Inspired by this, we propose LRF, a novel \textbf{L}ayer-wise \textbf{R}epresentation \textbf{F}usion framework for CG, which learns to fuse previous layers' information back into the encoding and decoding process effectively through introducing a \emph{fuse-attention module} at each encoder and decoder layer. LRF achieves promising results on two realistic benchmarks, empirically demonstrating the effectiveness of our proposal.
Global single-valued biomarkers of cardiac function typically used in clinical practice, such as ejection fraction, provide limited insight on the true 3D cardiac deformation process and hence, limit the understanding of both healthy and pathological cardiac mechanics. In this work, we propose the Point Cloud Deformation Network (PCD-Net) as a novel geometric deep learning approach to model 3D cardiac contraction and relaxation between the extreme ends of the cardiac cycle. It employs the recent advances in point cloud-based deep learning into an encoder-decoder structure, in order to enable efficient multi-scale feature learning directly on multi-class 3D point cloud representations of the cardiac anatomy. We evaluate our approach on a large dataset of over 10,000 cases from the UK Biobank study and find average Chamfer distances between the predicted and ground truth anatomies below the pixel resolution of the underlying image acquisition. Furthermore, we observe similar clinical metrics between predicted and ground truth populations and show that the PCD-Net can successfully capture subpopulation-specific differences between normal subjects and myocardial infarction (MI) patients. We then demonstrate that the learned 3D deformation patterns outperform multiple clinical benchmarks by 13% and 7% in terms of area under the receiver operating characteristic curve for the tasks of prevalent MI detection and incident MI prediction and by 7% in terms of Harrell's concordance index for MI survival analysis.
Weakly-supervised change detection (WSCD) aims to detect pixel-level changes with only image-level annotations. Owing to its label efficiency, WSCD is drawing increasing attention recently. However, current WSCD methods often encounter the challenge of change missing and fabricating, i.e., the inconsistency between image-level annotations and pixel-level predictions. Specifically, change missing refer to the situation that the WSCD model fails to predict any changed pixels, even though the image-level label indicates changed, and vice versa for change fabricating. To address this challenge, in this work, we leverage global-scale and local-scale priors in WSCD and propose two components: a Dilated Prior (DP) decoder and a Label Gated (LG) constraint. The DP decoder decodes samples with the changed image-level label, skips samples with the unchanged label, and replaces them with an all-unchanged pixel-level label. The LG constraint is derived from the correspondence between changed representations and image-level labels, penalizing the model when it mispredicts the change status. Additionally, we develop TransWCD, a simple yet powerful transformer-based model, showcasing the potential of weakly-supervised learning in change detection. By integrating the DP decoder and LG constraint into TransWCD, we form TransWCD-DL. Our proposed TransWCD and TransWCD-DL achieve significant +6.33% and +9.55% F1 score improvements over the state-of-the-art methods on the WHU-CD dataset, respectively. Some performance metrics even exceed several fully-supervised change detection (FSCD) competitors. Code will be available at https://github.com/zhenghuizhao/TransWCD.
Building outline extracted from high-resolution aerial images can be used in various application fields such as change detection and disaster assessment. However, traditional CNN model cannot recognize contours very precisely from original images. In this paper, we proposed a CNN and Transformer based model together with active contour model to deal with this problem. We also designed a triple-branch decoder structure to handle different features generated by encoder. Experiment results show that our model outperforms other baseline model on two datasets, achieving 91.1% mIoU on Vaihingen and 83.8% on Bing huts.
Object localization in general environments is a fundamental part of vision systems. While dominating on the COCO benchmark, recent Transformer-based detection methods are not competitive in diverse domains. Moreover, these methods still struggle to very accurately estimate the object bounding boxes in complex environments. We introduce Cascade-DETR for high-quality universal object detection. We jointly tackle the generalization to diverse domains and localization accuracy by proposing the Cascade Attention layer, which explicitly integrates object-centric information into the detection decoder by limiting the attention to the previous box prediction. To further enhance accuracy, we also revisit the scoring of queries. Instead of relying on classification scores, we predict the expected IoU of the query, leading to substantially more well-calibrated confidences. Lastly, we introduce a universal object detection benchmark, UDB10, that contains 10 datasets from diverse domains. While also advancing the state-of-the-art on COCO, Cascade-DETR substantially improves DETR-based detectors on all datasets in UDB10, even by over 10 mAP in some cases. The improvements under stringent quality requirements are even more pronounced. Our code and models will be released at https://github.com/SysCV/cascade-detr.
State-of-the-art text-video retrieval (TVR) methods typically utilize CLIP and cosine similarity for efficient retrieval. Meanwhile, cross attention methods, which employ a transformer decoder to compute attention between each text query and all frames in a video, offer a more comprehensive interaction between text and videos. However, these methods lack important fine-grained spatial information as they directly compute attention between text and video-level tokens. To address this issue, we propose CrossTVR, a two-stage text-video retrieval architecture. In the first stage, we leverage existing TVR methods with cosine similarity network for efficient text/video candidate selection. In the second stage, we propose a novel decoupled video text cross attention module to capture fine-grained multimodal information in spatial and temporal dimensions. Additionally, we employ the frozen CLIP model strategy in fine-grained retrieval, enabling scalability to larger pre-trained vision models like ViT-G, resulting in improved retrieval performance. Experiments on text video retrieval datasets demonstrate the effectiveness and scalability of our proposed CrossTVR compared to state-of-the-art approaches.
We view variational autoencoders (VAE) as decoder-encoder pairs, which map distributions in the data space to distributions in the latent space and vice versa. The standard learning approach for VAEs is the maximisation of the evidence lower bound (ELBO). It is asymmetric in that it aims at learning a latent variable model while using the encoder as an auxiliary means only. Moreover, it requires a closed form a-priori latent distribution. This limits its applicability in more complex scenarios, such as general semi-supervised learning and employing complex generative models as priors. We propose a Nash equilibrium learning approach, which is symmetric with respect to the encoder and decoder and allows learning VAEs in situations where both the data and the latent distributions are accessible only by sampling. The flexibility and simplicity of this approach allows its application to a wide range of learning scenarios and downstream tasks.
In dyadic speaker-listener interactions, the listener's head reactions along with the speaker's head movements, constitute an important non-verbal semantic expression together. The listener Head generation task aims to synthesize responsive listener's head videos based on audios of the speaker and reference images of the listener. Compared to the Talking-head generation, it is more challenging to capture the correlation clues from the speaker's audio and visual information. Following the ViCo baseline scheme, we propose a high-performance solution by enhancing the hierarchical semantic extraction capability of the audio encoder module and improving the decoder part, renderer and post-processing modules. Our solution gets the first place on the official leaderboard for the track of listening head generation. This paper is a technical report of ViCo@2023 Conversational Head Generation Challenge in ACM Multimedia 2023 conference.
Neural decoders for quantum error correction (QEC) rely on neural networks to classify syndromes extracted from error correction codes and find appropriate recovery operators to protect logical information against errors. Its ability to adapt to hardware noise and long-term drifts make neural decoders a promising candidate for inclusion in a fault-tolerant quantum architecture. However, given their limited scalability, it is prudent that small-scale (local) neural decoders are treated as first stages of multi-stage decoding schemes for fault-tolerant quantum computers with millions of qubits. In this case, minimizing the decoding time to match the stabilization measurements frequency and a tight co-integration with the QPUs is highly desired. Cryogenic realizations of neural decoders can not only improve the performance of higher stage decoders, but they can minimize communication delays, and alleviate wiring bottlenecks. In this work, we design and analyze a neural decoder based on an in-memory computation (IMC) architecture, where crossbar arrays of resistive memory devices are employed to both store the synaptic weights of the neural decoder and perform analog matrix-vector multiplications. In simulations supported by experimental measurements, we investigate the impact of TiOx-based memristive devices' non-idealities on decoding fidelity. We develop hardware-aware re-training methods to mitigate the fidelity loss, restoring the ideal decoder's pseudo-threshold for the distance-3 surface code. This work provides a pathway to scalable, fast, and low-power cryogenic IMC hardware for integrated fault-tolerant QEC.
Microseismic event detection and location are two primary components in microseismic monitoring, which offers us invaluable insights into the subsurface during reservoir stimulation and evolution. Conventional approaches for event detection and location often suffer from manual intervention and/or heavy computation, while current machine learning-assisted approaches typically address detection and location separately; such limitations hinder the potential for real-time microseismic monitoring. We propose an approach to unify event detection and source location into a single framework by adapting a Convolutional Neural Network backbone and an encoder-decoder Transformer with a set-based Hungarian loss, which is applied directly to recorded waveforms. The proposed network is trained on synthetic data simulating multiple microseismic events corresponding to random source locations in the area of suspected microseismic activities. A synthetic test on a 2D profile of the SEAM Time Lapse model illustrates the capability of the proposed method in detecting the events properly and locating them in the subsurface accurately; while, a field test using the Arkoma Basin data further proves its practicability, efficiency, and its potential in paving the way for real-time monitoring of microseismic events.
For problems in image processing and many other fields, a large class of effective neural networks has encoder-decoder-based architectures. Although these networks have made impressive performances, mathematical explanations of their architectures are still underdeveloped. In this paper, we study the encoder-decoder-based network architecture from the algorithmic perspective and provide a mathematical explanation. We use the two-phase Potts model for image segmentation as an example for our explanations. We associate the segmentation problem with a control problem in the continuous setting. Then, multigrid method and operator splitting scheme, the PottsMGNet, are used to discretize the continuous control model. We show that the resulting discrete PottsMGNet is equivalent to an encoder-decoder-based network. With minor modifications, it is shown that a number of the popular encoder-decoder-based neural networks are just instances of the proposed PottsMGNet. By incorporating the Soft-Threshold-Dynamics into the PottsMGNet as a regularizer, the PottsMGNet has shown to be robust with the network parameters such as network width and depth and achieved remarkable performance on datasets with very large noise. In nearly all our experiments, the new network always performs better or as good on accuracy and dice score than existing networks for image segmentation.
Accurate and reliable ego-localization is critical for autonomous driving. In this paper, we present EgoVM, an end-to-end localization network that achieves comparable localization accuracy to prior state-of-the-art methods, but uses lightweight vectorized maps instead of heavy point-based maps. To begin with, we extract BEV features from online multi-view images and LiDAR point cloud. Then, we employ a set of learnable semantic embeddings to encode the semantic types of map elements and supervise them with semantic segmentation, to make their feature representation consistent with BEV features. After that, we feed map queries, composed of learnable semantic embeddings and coordinates of map elements, into a transformer decoder to perform cross-modality matching with BEV features. Finally, we adopt a robust histogram-based pose solver to estimate the optimal pose by searching exhaustively over candidate poses. We comprehensively validate the effectiveness of our method using both the nuScenes dataset and a newly collected dataset. The experimental results show that our method achieves centimeter-level localization accuracy, and outperforms existing methods using vectorized maps by a large margin. Furthermore, our model has been extensively tested in a large fleet of autonomous vehicles under various challenging urban scenes.
Most end-to-end (E2E) speech recognition models are composed of encoder and decoder blocks that perform acoustic and language modeling functions. Pretrained large language models (LLMs) have the potential to improve the performance of E2E ASR. However, integrating a pretrained language model into an E2E speech recognition model has shown limited benefits due to the mismatches between text-based LLMs and those used in E2E ASR. In this paper, we explore an alternative approach by adapting a pretrained LLMs to speech. Our experiments on fully-formatted E2E ASR transcription tasks across various domains demonstrate that our approach can effectively leverage the strengths of pretrained LLMs to produce more readable ASR transcriptions. Our model, which is based on the pretrained large language models with either an encoder-decoder or decoder-only structure, surpasses strong ASR models such as Whisper, in terms of recognition error rate, considering formats like punctuation and capitalization as well.
Syntax has been proven to be remarkably effective in neural machine translation (NMT). Previous models obtained syntax information from syntactic parsing tools and integrated it into NMT models to improve translation performance. In this work, we propose a method to incorporate syntax information into a complex-valued Encoder-Decoder architecture. The proposed model jointly learns word-level and syntax-level attention scores from the source side to the target side using an attention mechanism. Importantly, it is not dependent on specific network architectures and can be directly integrated into any existing sequence-to-sequence (Seq2Seq) framework. The experimental results demonstrate that the proposed method can bring significant improvements in BLEU scores on two datasets. In particular, the proposed method achieves a greater improvement in BLEU scores in translation tasks involving language pairs with significant syntactic differences.
Optical information processing and computing can potentially offer enhanced performance, scalability and energy efficiency. However, achieving nonlinearity-a critical component of computation-remains challenging in the optical domain. Here we introduce a design that leverages a multiple-scattering cavity to passively induce optical nonlinear random mapping with a continuous-wave laser at a low power. Each scattering event effectively mixes information from different areas of a spatial light modulator, resulting in a highly nonlinear mapping between the input data and output pattern. We demonstrate that our design retains vital information even when the readout dimensionality is reduced, thereby enabling optical data compression. This capability allows our optical platforms to offer efficient optical information processing solutions across applications. We demonstrate our design's efficacy across tasks, including classification, image reconstruction, keypoint detection and object detection, all of which are achieved through optical data compression combined with a digital decoder. In particular, high performance at extreme compression ratios is observed in real-time pedestrian detection. Our findings open pathways for novel algorithms and unconventional architectural designs for optical computing.
This paper shows that time series forecasting Transformer (TSFT) suffers from severe over-fitting problem caused by improper initialization method of unknown decoder inputs, esp. when handling non-stationary time series. Based on this observation, we propose GBT, a novel two-stage Transformer framework with Good Beginning. It decouples the prediction process of TSFT into two stages, including Auto-Regression stage and Self-Regression stage to tackle the problem of different statistical properties between input and prediction sequences.Prediction results of Auto-Regression stage serve as a Good Beginning, i.e., a better initialization for inputs of Self-Regression stage. We also propose Error Score Modification module to further enhance the forecasting capability of the Self-Regression stage in GBT. Extensive experiments on seven benchmark datasets demonstrate that GBT outperforms SOTA TSFTs (FEDformer, Pyraformer, ETSformer, etc.) and many other forecasting models (SCINet, N-HiTS, etc.) with only canonical attention and convolution while owning less time and space complexity. It is also general enough to couple with these models to strengthen their forecasting capability. The source code is available at: https://github.com/OrigamiSL/GBT
In generative modeling, numerous successful approaches leverage a low-dimensional latent space, e.g., Stable Diffusion models the latent space induced by an encoder and generates images through a paired decoder. Although the selection of the latent space is empirically pivotal, determining the optimal choice and the process of identifying it remain unclear. In this study, we aim to shed light on this under-explored topic by rethinking the latent space from the perspective of model complexity. Our investigation starts with the classic generative adversarial networks (GANs). Inspired by the GAN training objective, we propose a novel "distance" between the latent and data distributions, whose minimization coincides with that of the generator complexity. The minimizer of this distance is characterized as the optimal data-dependent latent that most effectively capitalizes on the generator's capacity. Then, we consider parameterizing such a latent distribution by an encoder network and propose a two-stage training strategy called Decoupled Autoencoder (DAE), where the encoder is only updated in the first stage with an auxiliary decoder and then frozen in the second stage while the actual decoder is being trained. DAE can improve the latent distribution and as a result, improve the generative performance. Our theoretical analyses are corroborated by comprehensive experiments on various models such as VQGAN and Diffusion Transformer, where our modifications yield significant improvements in sample quality with decreased model complexity.
We propose a self-supervised deep learning-based decoding scheme that enables one-shot decoding of polar codes. In the proposed scheme, rather than using the information bit vectors as labels for training the neural network (NN) through supervised learning as the conventional scheme did, the NN is trained to function as a bounded distance decoder by leveraging the generator matrix of polar codes through self-supervised learning. This approach eliminates the reliance on predefined labels, empowering the potential to train directly on the actual data within communication systems and thereby enhancing the applicability. Furthermore, computer simulations demonstrate that (i) the bit error rate (BER) and block error rate (BLER) performances of the proposed scheme can approach those of the maximum a posteriori (MAP) decoder for very short packets and (ii) the proposed NN decoder (NND) exhibits much superior generalization ability compared to the conventional one.
Accurately predicting deformable linear object (DLO) dynamics is challenging, especially when the task requires a model that is both human-interpretable and computationally efficient. In this work, we draw inspiration from the pseudo-rigid body method (PRB) and model a DLO as a serial chain of rigid bodies whose internal state is unrolled through time by a dynamics network. This dynamics network is trained jointly with a physics-informed encoder that maps observed motion variables to the DLO's hidden state. To encourage the state to acquire a physically meaningful representation, we leverage the forward kinematics of the PRB model as a decoder. We demonstrate in robot experiments that the proposed DLO dynamics model provides physically interpretable predictions from partial observations while being on par with black-box models regarding prediction accuracy. The project code is available at: http://tinyurl.com/prb-networks