Organ at risk (OAR) segmentation in computed tomography (CT) imagery is a difficult task for automated segmentation methods and can be crucial for downstream radiation treatment planning. U-net has become a de-facto standard for medical image segmentation and is frequently used as a common baseline in medical image segmentation tasks. In this paper, we propose a multiple decoder U-net architecture and use the segmentation disagreement between the decoders as attention to the bottleneck of the network for segmentation refinement. While feature correlation is considered as attention in most cases, in our case it is the uncertainty from the network used as attention. For accurate segmentation, we also proposed a CT intensity integrated regularization loss. Proposed regularisation helps model understand the intensity distribution of low contrast tissues. We tested our model on two publicly available OAR challenge datasets. We also conducted the ablation on each datasets with the proposed attention module and regularization loss. Experimental results demonstrate a clear accuracy improvement on both datasets.
Point cloud segmentation is a fundamental task in 3D vision that serves a wide range of applications. Although great progresses have been made these years, its practical usability is still limited by the availability of training data. Existing approaches cannot make full use of multiple datasets on hand due to the label mismatch among different datasets. In this paper, we propose a principled approach that supports learning from heterogeneous datasets with different label sets. Our idea is to utilize a pre-trained language model to embed discrete labels to a continuous latent space with the help of their label names. This unifies all labels of different datasets, so that joint training is doable. Meanwhile, classifying points in the continuous 3D space by their vocabulary tokens significantly increase the generalization ability of the model in comparison with existing approaches that have fixed decoder architecture. Besides, we also integrate prompt learning in our framework to alleviate data shifts among different data sources. Extensive experiments demonstrate that our model outperforms the state-of-the-art by a large margin.
Object segmentation for robotic grasping under dynamic conditions often faces challenges such as occlusion, low light conditions, motion blur and object size variance. To address these challenges, we propose a Deep Learning network that fuses two types of visual signals, event-based data and RGB frame data. The proposed Bimodal SegNet network has two distinct encoders, one for each signal input and a spatial pyramidal pooling with atrous convolutions. Encoders capture rich contextual information by pooling the concatenated features at different resolutions while the decoder obtains sharp object boundaries. The evaluation of the proposed method undertakes five unique image degradation challenges including occlusion, blur, brightness, trajectory and scale variance on the Event-based Segmentation (ESD) Dataset. The evaluation results show a 6-10\% segmentation accuracy improvement over state-of-the-art methods in terms of mean intersection over the union and pixel accuracy. The model code is available at https://github.com/sanket0707/Bimodal-SegNet.git
Given an abstract, deformed, ordinary sketch from untrained amateurs like you and me, this paper turns it into a photorealistic image - just like those shown in Fig. 1(a), all non-cherry-picked. We differ significantly from prior art in that we do not dictate an edgemap-like sketch to start with, but aim to work with abstract free-hand human sketches. In doing so, we essentially democratise the sketch-to-photo pipeline, "picturing" a sketch regardless of how good you sketch. Our contribution at the outset is a decoupled encoder-decoder training paradigm, where the decoder is a StyleGAN trained on photos only. This importantly ensures that generated results are always photorealistic. The rest is then all centred around how best to deal with the abstraction gap between sketch and photo. For that, we propose an autoregressive sketch mapper trained on sketch-photo pairs that maps a sketch to the StyleGAN latent space. We further introduce specific designs to tackle the abstract nature of human sketches, including a fine-grained discriminative loss on the back of a trained sketch-photo retrieval model, and a partial-aware sketch augmentation strategy. Finally, we showcase a few downstream tasks our generation model enables, amongst them is showing how fine-grained sketch-based image retrieval, a well-studied problem in the sketch community, can be reduced to an image (generated) to image retrieval task, surpassing state-of-the-arts. We put forward generated results in the supplementary for everyone to scrutinise.
Speech-driven 3D face animation aims to generate realistic facial expressions that match the speech content and emotion. However, existing methods often neglect emotional facial expressions or fail to disentangle them from speech content. To address this issue, this paper proposes an end-to-end neural network to disentangle different emotions in speech so as to generate rich 3D facial expressions. Specifically, we introduce the emotion disentangling encoder (EDE) to disentangle the emotion and content in the speech by cross-reconstructed speech signals with different emotion labels. Then an emotion-guided feature fusion decoder is employed to generate a 3D talking face with enhanced emotion. The decoder is driven by the disentangled identity, emotional, and content embeddings so as to generate controllable personal and emotional styles. Finally, considering the scarcity of the 3D emotional talking face data, we resort to the supervision of facial blendshapes, which enables the reconstruction of plausible 3D faces from 2D emotional data, and contribute a large-scale 3D emotional talking face dataset (3D-ETF) to train the network. Our experiments and user studies demonstrate that our approach outperforms state-of-the-art methods and exhibits more diverse facial movements. We recommend watching the supplementary video: https://ziqiaopeng.github.io/emotalk
Semantic segmentation is a classic and fundamental computer vision problem dedicated to assigning each pixel with its corresponding class. Some recent methods introduce edge-based information for improving the segmentation performance. However these methods are specific and limited to certain network architectures, and they can not be transferred to other models or tasks. Therefore, we propose an abstract and universal edge supervision method called Edge-aware Plug-and-play Scheme (EPS), which can be easily and quickly applied to any semantic segmentation models. The core is edge-width/thickness preserving guided for semantic segmentation. The EPS first extracts the Edge Ground Truth (Edge GT) with a predefined edge thickness from the training data; and then for any network architecture, it directly copies the decoder head for the auxiliary task with the Edge GT supervision. To ensure the edge thickness preserving consistantly, we design a new boundarybased loss, called Polar Hausdorff (PH) Loss, for the auxiliary supervision. We verify the effectiveness of our EPS on the Cityscapes dataset using 22 models. The experimental results indicate that the proposed method can be seamlessly integrated into any state-of-the-art (SOTA) models with zero modification, resulting in promising enhancement of the segmentation performance.
Multi-view camera-based 3D detection is a challenging problem in computer vision. Recent works leverage a pretrained LiDAR detection model to transfer knowledge to a camera-based student network. However, we argue that there is a major domain gap between the LiDAR BEV features and the camera-based BEV features, as they have different characteristics and are derived from different sources. In this paper, we propose Geometry Enhanced Masked Image Modeling (GeoMIM) to transfer the knowledge of the LiDAR model in a pretrain-finetune paradigm for improving the multi-view camera-based 3D detection. GeoMIM is a multi-camera vision transformer with Cross-View Attention (CVA) blocks that uses LiDAR BEV features encoded by the pretrained BEV model as learning targets. During pretraining, GeoMIM's decoder has a semantic branch completing dense perspective-view features and the other geometry branch reconstructing dense perspective-view depth maps. The depth branch is designed to be camera-aware by inputting the camera's parameters for better transfer capability. Extensive results demonstrate that GeoMIM outperforms existing methods on nuScenes benchmark, achieving state-of-the-art performance for camera-based 3D object detection and 3D segmentation. Code and pretrained models are available at https://github.com/Sense-X/GeoMIM.
Background: In medical imaging, images are usually treated as deterministic, while their uncertainties are largely underexplored. Purpose: This work aims at using deep learning to efficiently estimate posterior distributions of imaging parameters, which in turn can be used to derive the most probable parameters as well as their uncertainties. Methods: Our deep learning-based approaches are based on a variational Bayesian inference framework, which is implemented using two different deep neural networks based on conditional variational auto-encoder (CVAE), CVAE-dual-encoder and CVAE-dual-decoder. The conventional CVAE framework, i.e., CVAE-vanilla, can be regarded as a simplified case of these two neural networks. We applied these approaches to a simulation study of dynamic brain PET imaging using a reference region-based kinetic model. Results: In the simulation study, we estimated posterior distributions of PET kinetic parameters given a measurement of time-activity curve. Our proposed CVAE-dual-encoder and CVAE-dual-decoder yield results that are in good agreement with the asymptotically unbiased posterior distributions sampled by Markov Chain Monte Carlo (MCMC). The CVAE-vanilla can also be used for estimating posterior distributions, although it has an inferior performance to both CVAE-dual-encoder and CVAE-dual-decoder. Conclusions: We have evaluated the performance of our deep learning approaches for estimating posterior distributions in dynamic brain PET. Our deep learning approaches yield posterior distributions, which are in good agreement with unbiased distributions estimated by MCMC. All these neural networks have different characteristics and can be chosen by the user for specific applications. The proposed methods are general and can be adapted to other problems.
Text-to-image (T2I) models based on diffusion processes have achieved remarkable success in controllable image generation using user-provided captions. However, the tight coupling between the current text encoder and image decoder in T2I models makes it challenging to replace or upgrade. Such changes often require massive fine-tuning or even training from scratch with the prohibitive expense. To address this problem, we propose GlueGen, which applies a newly proposed GlueNet model to align features from single-modal or multi-modal encoders with the latent space of an existing T2I model. The approach introduces a new training objective that leverages parallel corpora to align the representation spaces of different encoders. Empirical results show that GlueNet can be trained efficiently and enables various capabilities beyond previous state-of-the-art models: 1) multilingual language models such as XLM-Roberta can be aligned with existing T2I models, allowing for the generation of high-quality images from captions beyond English; 2) GlueNet can align multi-modal encoders such as AudioCLIP with the Stable Diffusion model, enabling sound-to-image generation; 3) it can also upgrade the current text encoder of the latent diffusion model for challenging case generation. By the alignment of various feature representations, the GlueNet allows for flexible and efficient integration of new functionality into existing T2I models and sheds light on X-to-image (X2I) generation.
Classification and segmentation are crucial in medical image analysis as they enable accurate diagnosis and disease monitoring. However, current methods often prioritize the mutual learning features and shared model parameters, while neglecting the reliability of features and performances. In this paper, we propose a novel Uncertainty-informed Mutual Learning (UML) framework for reliable and interpretable medical image analysis. Our UML introduces reliability to joint classification and segmentation tasks, leveraging mutual learning with uncertainty to improve performance. To achieve this, we first use evidential deep learning to provide image-level and pixel-wise confidences. Then, an Uncertainty Navigator Decoder is constructed for better using mutual features and generating segmentation results. Besides, an Uncertainty Instructor is proposed to screen reliable masks for classification. Overall, UML could produce confidence estimation in features and performance for each link (classification and segmentation). The experiments on the public datasets demonstrate that our UML outperforms existing methods in terms of both accuracy and robustness. Our UML has the potential to explore the development of more reliable and explainable medical image analysis models. We will release the codes for reproduction after acceptance.
There has been exploding interest in embracing Transformer-based architectures for medical image segmentation. However, the lack of large-scale annotated medical datasets make achieving performances equivalent to those in natural images challenging. Convolutional networks, in contrast, have higher inductive biases and consequently, are easily trainable to high performance. Recently, the ConvNeXt architecture attempted to modernize the standard ConvNet by mirroring Transformer blocks. In this work, we improve upon this to design a modernized and scalable convolutional architecture customized to challenges of data-scarce medical settings. We introduce MedNeXt, a Transformer-inspired large kernel segmentation network which introduces - 1) A fully ConvNeXt 3D Encoder-Decoder Network for medical image segmentation, 2) Residual ConvNeXt up and downsampling blocks to preserve semantic richness across scales, 3) A novel technique to iteratively increase kernel sizes by upsampling small kernel networks, to prevent performance saturation on limited medical data, 4) Compound scaling at multiple levels (depth, width, kernel size) of MedNeXt. This leads to state-of-the-art performance on 4 tasks on CT and MRI modalities and varying dataset sizes, representing a modernized deep architecture for medical image segmentation. Our code is made publicly available at: https://github.com/MIC-DKFZ/MedNeXt.
Visual-Language Models (VLMs) have significantly advanced action video recognition. Supervised by the semantics of action labels, recent works adapt the visual branch of VLMs to learn video representations. Despite the effectiveness proved by these works, we believe that the potential of VLMs has yet to be fully harnessed. In light of this, we exploit the semantic units (SU) hiding behind the action labels and leverage their correlations with fine-grained items in frames for more accurate action recognition. SUs are entities extracted from the language descriptions of the entire action set, including body parts, objects, scenes, and motions. To further enhance the alignments between visual contents and the SUs, we introduce a multi-region module (MRA) to the visual branch of the VLM. The MRA allows the perception of region-aware visual features beyond the original global feature. Our method adaptively attends to and selects relevant SUs with visual features of frames. With a cross-modal decoder, the selected SUs serve to decode spatiotemporal video representations. In summary, the SUs as the medium can boost discriminative ability and transferability. Specifically, in fully-supervised learning, our method achieved 87.8% top-1 accuracy on Kinetics-400. In K=2 few-shot experiments, our method surpassed the previous state-of-the-art by +7.1% and +15.0% on HMDB-51 and UCF-101, respectively.
Garment pattern design aims to convert a 3D garment to the corresponding 2D panels and their sewing structure. Existing methods rely either on template fitting with heuristics and prior assumptions, or on model learning with complicated shape parameterization. Importantly, both approaches do not allow for personalization of the output garment, which today has increasing demands. To fill this demand, we introduce PersonalTailor: a personalized 2D pattern design method, where the user can input specific constraints or demands (in language or sketch) for personal 2D panel fabrication from 3D point clouds. PersonalTailor first learns a multi-modal panel embeddings based on unsupervised cross-modal association and attentive fusion. It then predicts a binary panel masks individually using a transformer encoder-decoder framework. Extensive experiments show that our PersonalTailor excels on both personalized and standard pattern fabrication tasks.
Semantic segmentation is a fundamental capability for autonomous vehicles. With the advancements of deep learning technologies, many effective semantic segmentation networks have been proposed in recent years. However, most of them are designed using RGB images from visible cameras. The quality of RGB images is prone to be degraded under unsatisfied lighting conditions, such as darkness and glares of oncoming headlights, which imposes critical challenges for the networks that use only RGB images. Different from visible cameras, thermal imaging cameras generate images using thermal radiations. They are able to see under various lighting conditions. In order to enable robust and accurate semantic segmentation for autonomous vehicles, we take the advantage of thermal images and fuse both the RGB and thermal information in a novel deep neural network. The main innovation of this letter is the architecture of the proposed network. We adopt the encoder–decoder design concept. ResNet is employed for feature extraction and a new decoder is developed to restore the feature map resolution. The experimental results prove that our network outperforms the state of the arts.
Zero-shot learning (ZSL) aims to recognize unseen classes by generalizing the relation between visual features and semantic attributes learned from the seen classes. A recent paradigm called transductive zero-shot learning further leverages unlabeled unseen data during training and has obtained impressive results. These methods always synthesize unseen features from attributes through a generative adversarial network to mitigate the bias towards seen classes. However, they neglect the semantic information in the unlabeled unseen data and thus fail to generate high-fidelity attribute-consistent unseen features. To address this issue, we present a novel transductive ZSL method that produces semantic attributes of the unseen data and imposes them on the generative process. In particular, we first train an attribute decoder that learns the mapping from visual features to semantic attributes. Then, from the attribute decoder, we obtain pseudo-attributes of unlabeled data and integrate them into the generative model, which helps capture the detailed differences within unseen classes so as to synthesize more discriminative features. Experiments on five standard benchmarks show that our method yields state-of-the-art results for zero-shot learning.
In edge intelligence, deep learning~(DL) models are deployed at an edge device and an edge server for data processing with low latency in the Internet of Things~(IoT). In this letter, we propose a new end-to-end learning-based wireless image recognition scheme using the PyramidNet in edge intelligence. We split the PyramidNet carefully into two parts for an IoT device and the edge server, which is to pursue low on-device computation. Also, we apply a squeeze-and-excitation block to the PyramidNet for the improvement of image recognition. In addition, we embed compression encoder and decoder at the splitting point, which reduces communication overhead by compressing the intermediate feature map. Simulation results demonstrate that the proposed scheme is superior to other DL-based schemes in image recognition, while presenting less on-device computation and fewer parameters with low communication overhead.
Voice Conversion (VC) must be achieved while maintaining the content of the source speech and representing the characteristics of the target speaker. The existing methods do not simultaneously satisfy the above two aspects of VC, and their conversion outputs suffer from a trade-off problem between maintaining source contents and target characteristics. In this study, we propose Triple Adaptive Attention Normalization VC (TriAAN-VC), comprising an encoder-decoder and an attention-based adaptive normalization block, that can be applied to non-parallel any-to-any VC. The proposed adaptive normalization block extracts target speaker representations and achieves conversion while minimizing the loss of the source content with siamese loss. We evaluated TriAAN-VC on the VCTK dataset in terms of the maintenance of the source content and target speaker similarity. Experimental results for one-shot VC suggest that TriAAN-VC achieves state-of-the-art performance while mitigating the trade-off problem encountered in the existing VC methods.
On a daily basis, data centers process huge volumes of data backed by the proliferation of inexpensive hard disks. Data stored in these disks serve a range of critical functional needs from financial, and healthcare to aerospace. As such, premature disk failure and consequent loss of data can be catastrophic. To mitigate the risk of failures, cloud storage providers perform condition-based monitoring and replace hard disks before they fail. By estimating the remaining useful life of hard disk drives, one can predict the time-to-failure of a particular device and replace it at the right time, ensuring maximum utilization whilst reducing operational costs. In this work, large-scale predictive analyses are performed using severely skewed health statistics data by incorporating customized feature engineering and a suite of sequence learners. Past work suggests using LSTMs as an excellent approach to predicting remaining useful life. To this end, we present an encoder-decoder LSTM model where the context gained from understanding health statistics sequences aid in predicting an output sequence of the number of days remaining before a disk potentially fails. The models developed in this work are trained and tested across an exhaustive set of all of the 10 years of S.M.A.R.T. health data in circulation from Backblaze and on a wide variety of disk instances. It closes the knowledge gap on what full-scale training achieves on thousands of devices and advances the state-of-the-art by providing tangible metrics for evaluation and generalization for practitioners looking to extend their workflow to all years of health data in circulation across disk manufacturers. The encoder-decoder LSTM posted an RMSE of 0.83 during training and 0.86 during testing over the exhaustive 10 year data while being able to generalize competitively over other drives from the Seagate family.
We introduce an on-ground Pedestrian World Model, a computational model that can predict how pedestrians move around an observer in the crowd on the ground plane, but from just the egocentric-views of the observer. Our model, InCrowdFormer, fully leverages the Transformer architecture by modeling pedestrian interaction and egocentric to top-down view transformation with attention, and autoregressively predicts on-ground positions of a variable number of people with an encoder-decoder architecture. We encode the uncertainties arising from unknown pedestrian heights with latent codes to predict the posterior distributions of pedestrian positions. We validate the effectiveness of InCrowdFormer on a novel prediction benchmark of real movements. The results show that InCrowdFormer accurately predicts the future coordination of pedestrians. To the best of our knowledge, InCrowdFormer is the first-of-its-kind pedestrian world model which we believe will benefit a wide range of egocentric-view applications including crowd navigation, tracking, and synthesis.
Noise synthesis is a challenging low-level vision task aiming to generate realistic noise given a clean image along with the camera settings. To this end, we propose an effective generative model which utilizes clean features as guidance followed by noise injections into the network. Specifically, our generator follows a UNet-like structure with skip connections but without downsampling and upsampling layers. Firstly, we extract deep features from a clean image as the guidance and concatenate a Gaussian noise map to the transition point between the encoder and decoder as the noise source. Secondly, we propose noise synthesis blocks in the decoder in each of which we inject Gaussian noise to model the noise characteristics. Thirdly, we propose to utilize an additional Style Loss and demonstrate that this allows better noise characteristics supervision in the generator. Through a number of new experiments, we evaluate the temporal variance and the spatial correlation of the generated noise which we hope can provide meaningful insights for future works. Finally, we show that our proposed approach outperforms existing methods for synthesizing camera noise.