Qinfeng Shi

Hongrong Cheng, Miao Zhang, Javen Qinfeng Shi

Modern deep neural networks, particularly recent large language models, come with massive model sizes that require significant computational and storage resources. To enable the deployment of modern models on resource-constrained environments and accelerate inference time, researchers have increasingly explored pruning techniques as a popular research direction in neural network compression. However, there is a dearth of up-to-date comprehensive review papers on pruning. To address this issue, in this survey, we provide a comprehensive review of existing research works on deep neural network pruning in a taxonomy of 1) universal/specific speedup, 2) when to prune, 3) how to prune, and 4) fusion of pruning and other compression techniques. We then provide a thorough comparative analysis of eight pairs of contrast settings for pruning and explore emerging topics, including pruning for large language models, large multimodal models, post-training pruning, and different supervision levels for pruning to shed light on the commonalities and differences of existing methods and lay the foundation for further method development. To facilitate future research, we build a curated collection of datasets, networks, and evaluations on different applications. Finally, we provide valuable recommendations on selecting pruning methods and prospect several promising research directions. We build a repository at https://github.com/hrcheng1066/awesome-pruning.

Xinyu Zhang, Dongdong Li, Zhigang Wang et al.

Most existing unsupervised person re-identification (Re-ID) methods use clustering to generate pseudo labels for model training. Unfortunately, clustering sometimes mixes different true identities together or splits the same identity into two or more sub clusters. Training on these noisy clusters substantially hampers the Re-ID accuracy. Due to the limited samples in each identity, we suppose there may lack some underlying information to well reveal the accurate clusters. To discover these information, we propose an Implicit Sample Extension (\OurWholeMethod) method to generate what we call support samples around the cluster boundaries. Specifically, we generate support samples from actual samples and their neighbouring clusters in the embedding space through a progressive linear interpolation (PLI) strategy. PLI controls the generation with two critical factors, i.e., 1) the direction from the actual sample towards its K-nearest clusters and 2) the degree for mixing up the context information from the K-nearest clusters. Meanwhile, given the support samples, ISE further uses a label-preserving loss to pull them towards their corresponding actual samples, so as to compact each cluster. Consequently, ISE reduces the "sub and mixed" clustering errors, thus improving the Re-ID performance. Extensive experiments demonstrate that the proposed method is effective and achieves state-of-the-art performance for unsupervised person Re-ID. Code is available at: \url{https://github.com/PaddlePaddle/PaddleClas}.

Jinan Zou, Haiyao Cao, Lingqiao Liu et al.

Natural Language Processing(NLP) demonstrates a great potential to support financial decision-making by analyzing the text from social media or news outlets. In this work, we build a platform to study the NLP-aided stock auto-trading algorithms systematically. In contrast to the previous work, our platform is characterized by three features: (1) We provide financial news for each specific stock. (2) We provide various stock factors for each stock. (3) We evaluate performance from more financial-relevant metrics. Such a design allows us to develop and evaluate NLP-aided stock auto-trading algorithms in a more realistic setting. In addition to designing an evaluation platform and dataset collection, we also made a technical contribution by proposing a system to automatically learn a good feature representation from various input information. The key to our algorithm is a method called semantic role labeling Pooling (SRLP), which leverages Semantic Role Labeling (SRL) to create a compact representation of each news paragraph. Based on SRLP, we further incorporate other stock factors to make the final prediction. In addition, we propose a self-supervised learning strategy based on SRLP to enhance the out-of-distribution generalization performance of our system. Through our experimental study, we show that the proposed method achieves better performance and outperforms all the baselines' annualized rate of return as well as the maximum drawdown of the CSI300 index and XIN9 index on real trading. Our Astock dataset and code are available at https://github.com/JinanZou/Astock.

Hongrong Cheng, Miao Zhang, Javen Qinfeng Shi

A challenge of channel pruning is designing efficient and effective criteria to select channels to prune. A widely used criterion is minimal performance degeneration. To accurately evaluate the truth performance degeneration requires retraining the survived weights to convergence, which is prohibitively slow. Hence existing pruning methods use previous weights (without retraining) to evaluate the performance degeneration. However, we observe the loss changes differ significantly with and without retraining. It motivates us to develop a technique to evaluate true loss changes without retraining, with which channels to prune can be selected more reliably and confidently. We first derive a closed-form estimator of the true loss change per pruning mask change, using influence functions without retraining. Influence function which is from robust statistics reveals the impacts of a training sample on the model's prediction and is repurposed by us to assess impacts on true loss changes. We then show how to assess the importance of all channels simultaneously and develop a novel global channel pruning algorithm accordingly. We conduct extensive experiments to verify the effectiveness of the proposed algorithm. To the best of our knowledge, we are the first that shows evaluating true loss changes for pruning without retraining is possible. This finding will open up opportunities for a series of new paradigms to emerge that differ from existing pruning methods. The code is available at https://github.com/hrcheng1066/IFSO.

Amin Parvaneh, Ehsan Abbasnejad, Damien Teney et al.

The promise of active learning (AL) is to reduce labelling costs by selecting the most valuable examples to annotate from a pool of unlabelled data. Identifying these examples is especially challenging with high-dimensional data (e.g. images, videos) and in low-data regimes. In this paper, we propose a novel method for batch AL called ALFA-Mix. We identify unlabelled instances with sufficiently-distinct features by seeking inconsistencies in predictions resulting from interventions on their representations. We construct interpolations between representations of labelled and unlabelled instances then examine the predicted labels. We show that inconsistencies in these predictions help discovering features that the model is unable to recognise in the unlabelled instances. We derive an efficient implementation based on a closed-form solution to the optimal interpolation causing changes in predictions. Our method outperforms all recent AL approaches in 30 different settings on 12 benchmarks of images, videos, and non-visual data. The improvements are especially significant in low-data regimes and on self-trained vision transformers, where ALFA-Mix outperforms the state-of-the-art in 59% and 43% of the experiments respectively.

Eduardo Pérez-Pellitero, Sibi Catley-Chandar, Richard Shaw et al.

This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2022. This manuscript focuses on the competition set-up, datasets, the proposed methods and their results. The challenge aims at estimating an HDR image from multiple respective low dynamic range (LDR) observations, which might suffer from under- or over-exposed regions and different sources of noise. The challenge is composed of two tracks with an emphasis on fidelity and complexity constraints: In Track 1, participants are asked to optimize objective fidelity scores while imposing a low-complexity constraint (i.e. solutions can not exceed a given number of operations). In Track 2, participants are asked to minimize the complexity of their solutions while imposing a constraint on fidelity scores (i.e. solutions are required to obtain a higher fidelity score than the prescribed baseline). Both tracks use the same data and metrics: Fidelity is measured by means of PSNR with respect to a ground-truth HDR image (computed both directly and with a canonical tonemapping operation), while complexity metrics include the number of Multiply-Accumulate (MAC) operations and runtime (in seconds).

Yuhang Liu, Zhen Zhang, Dong Gong et al.

The task of causal representation learning aims to uncover latent higher-level causal variables that affect lower-level observations. Identifying the true latent causal variables from observed data, while allowing instantaneous causal relations among latent variables, remains a challenge, however. To this end, we start with the analysis of three intrinsic indeterminacies in identifying latent variables from observations: transitivity, permutation indeterminacy, and scaling indeterminacy. We find that transitivity acts as a key role in impeding the identifiability of latent causal variables. To address the unidentifiable issue due to transitivity, we introduce a novel identifiability condition where the underlying latent causal model satisfies a linear-Gaussian model, in which the causal coefficients and the distribution of Gaussian noise are modulated by an additional observed variable. Under certain assumptions, including the existence of a reference condition under which latent causal influences vanish, we can show that the latent causal variables can be identified up to trivial permutation and scaling, and that partial identifiability results can still be obtained when this reference condition is violated for a subset of latent variables. Furthermore, based on these theoretical results, we propose a novel method, termed Structural caUsAl Variational autoEncoder (SuaVE), which directly learns causal representations and causal relationships among them, together with the mapping from the latent causal variables to the observed ones. Experimental results on synthetic and real data demonstrate the identifiability and consistency results and the efficacy of SuaVE in learning causal representations.

Yuhang Liu, Zhen Zhang, Dong Gong et al.

Causal representation learning aims to unveil latent high-level causal representations from observed low-level data. One of its primary tasks is to provide reliable assurance of identifying these latent causal models, known as identifiability. A recent breakthrough explores identifiability by leveraging the change of causal influences among latent causal variables across multiple environments \citep{liu2022identifying}. However, this progress rests on the assumption that the causal relationships among latent causal variables adhere strictly to linear Gaussian models. In this paper, we extend the scope of latent causal models to involve nonlinear causal relationships, represented by polynomial models, and general noise distributions conforming to the exponential family. Additionally, we investigate the necessity of imposing changes on all causal parameters and present partial identifiability results when part of them remains unchanged. Further, we propose a novel empirical estimation method, grounded in our theoretical finding, that enables learning consistent latent causal representations. Our experimental results, obtained from both synthetic and real-world data, validate our theoretical contributions concerning identifiability and consistency.

Zhen Zhang, Ignavier Ng, Dong Gong et al.

Recovering underlying Directed Acyclic Graph (DAG) structures from observational data is highly challenging due to the combinatorial nature of the DAG-constrained optimization problem. Recently, DAG learning has been cast as a continuous optimization problem by characterizing the DAG constraint as a smooth equality one, generally based on polynomials over adjacency matrices. Existing methods place very small coefficients on high-order polynomial terms for stabilization, since they argue that large coefficients on the higher-order terms are harmful due to numeric exploding. On the contrary, we discover that large coefficients on higher-order terms are beneficial for DAG learning, when the spectral radiuses of the adjacency matrices are small, and that larger coefficients for higher-order terms can approximate the DAG constraints much better than the small counterparts. Based on this, we propose a novel DAG learning method with efficient truncated matrix power iteration to approximate geometric series based DAG constraints. Empirically, our DAG learning method outperforms the previous state-of-the-arts in various settings, often by a factor of $3$ or more in terms of structural Hamming distance.

Yichao Cai, Yuhang Liu, Zhen Zhang et al.

Contrastive vision-language models, such as CLIP, have garnered considerable attention for various downstream tasks, mainly due to the remarkable ability of the learned features for generalization. However, the features they learned often blend content and style information, which somewhat limits their generalization capabilities under distribution shifts. To address this limitation, we adopt a causal generative perspective for multimodal data and propose contrastive learning with data augmentation to disentangle content features from the original representations. To achieve this, we begin with exploring image augmentation techniques and develop a method to seamlessly integrate them into pre-trained CLIP-like models to extract pure content features. Taking a step further, recognizing the inherent semantic richness and logical structure of text data, we explore the use of text augmentation to isolate latent content from style features. This enables CLIP-like model's encoders to concentrate on latent content information, refining the learned representations by pre-trained CLIP-like models. Our extensive experiments across diverse datasets demonstrate significant improvements in zero-shot and few-shot classification tasks, alongside enhanced robustness to various perturbations. These results underscore the effectiveness of our proposed methods in refining vision-language representations and advancing the state-of-the-art in multimodal learning.

Yuhang Liu, Zhen Zhang, Dong Gong et al.

Multi-source domain adaptation (MSDA) addresses the challenge of learning a label prediction function for an unlabeled target domain by leveraging both the labeled data from multiple source domains and the unlabeled data from the target domain. Conventional MSDA approaches often rely on covariate shift or conditional shift paradigms, which assume a consistent label distribution across domains. However, this assumption proves limiting in practical scenarios where label distributions do vary across domains, diminishing its applicability in real-world settings. For example, animals from different regions exhibit diverse characteristics due to varying diets and genetics. Motivated by this, we propose a novel paradigm called latent covariate shift (LCS), which introduces significantly greater variability and adaptability across domains. Notably, it provides a theoretical assurance for recovering the latent cause of the label variable, which we refer to as the latent content variable. Within this new paradigm, we present an intricate causal generative model by introducing latent noises across domains, along with a latent content variable and a latent style variable to achieve more nuanced rendering of observational data. We demonstrate that the latent content variable can be identified up to block identifiability due to its versatile yet distinct causal structure. We anchor our theoretical insights into a novel MSDA method, which learns the label distribution conditioned on the identifiable latent content variable, thereby accommodating more substantial distribution shifts. The proposed approach showcases exceptional performance and efficacy on both simulated and real-world datasets.

Bao Gia Doan, Ehsan Abbasnejad, Javen Qinfeng Shi et al.

We present a new algorithm to learn a deep neural network model robust against adversarial attacks. Previous algorithms demonstrate an adversarially trained Bayesian Neural Network (BNN) provides improved robustness. We recognize the adversarial learning approach for approximating the multi-modal posterior distribution of a Bayesian model can lead to mode collapse; consequently, the model's achievements in robustness and performance are sub-optimal. Instead, we first propose preventing mode collapse to better approximate the multi-modal posterior distribution. Second, based on the intuition that a robust model should ignore perturbations and only consider the informative content of the input, we conceptualize and formulate an information gain objective to measure and force the information learned from both benign and adversarial training instances to be similar. Importantly. we prove and demonstrate that minimizing the information gain objective allows the adversarial risk to approach the conventional empirical risk. We believe our efforts provide a step toward a basis for a principled method of adversarially training BNNs. Our model demonstrate significantly improved robustness--up to 20%--compared with adversarial training and Adv-BNN under PGD attacks with 0.035 distortion on both CIFAR-10 and STL-10 datasets.

Haiyao Cao, Zhen Zhang, Panpan Cai et al.

One of the significant challenges in reinforcement learning (RL) when dealing with noise is estimating latent states from observations. Causality provides rigorous theoretical support for ensuring that the underlying states can be uniquely recovered through identifiability. Consequently, some existing work focuses on establishing identifiability from a causal perspective to aid in the design of algorithms. However, these results are often derived from a purely causal viewpoint, which may overlook the specific RL context. We revisit this research line and find that incorporating RL-specific context can reduce unnecessary assumptions in previous identifiability analyses for latent states. More importantly, removing these assumptions allows algorithm design to go beyond the earlier boundaries constrained by them. Leveraging these insights, we propose a novel approach for general partially observable Markov Decision Processes (POMDPs) by replacing the complicated structural constraints in previous methods with two simple constraints for transition and reward preservation. With the two constraints, the proposed algorithm is guaranteed to disentangle state and noise that is faithful to the underlying dynamics. Empirical evidence from extensive benchmark control tasks demonstrates the superiority of our approach over existing counterparts in effectively disentangling state belief from noise.

Hamed Damirchi, Cristian Rodríguez-Opazo, Ehsan Abbasnejad et al.

Large pre-trained models can dramatically reduce the amount of task-specific data required to solve a problem, but they often fail to capture domain-specific nuances out of the box. The Web likely contains the information necessary to excel on any specific application, but identifying the right data a priori is challenging. This paper shows how to leverage recent advances in NLP and multi-modal learning to augment a pre-trained model with search engine retrieval. We propose to retrieve useful data from the Web at test time based on test cases that the model is uncertain about. Different from existing retrieval-augmented approaches, we then update the model to address this underlying uncertainty. We demonstrate substantial improvements in zero-shot performance, e.g. a remarkable increase of 15 percentage points in accuracy on the Stanford Cars and Flowers datasets. We also present extensive experiments that explore the impact of noisy retrieval and different learning strategies.

Yuhao Lin, Haiming Xu, Lingqiao Liu et al.

Autoencoding, which aims to reconstruct the input images through a bottleneck latent representation, is one of the classic feature representation learning strategies. It has been shown effective as an auxiliary task for semi-supervised learning but has become less popular as more sophisticated methods have been proposed in recent years. In this paper, we revisit the idea of using image reconstruction as the auxiliary task and incorporate it with a modern semi-supervised semantic segmentation framework. Surprisingly, we discover that such an old idea in semi-supervised learning can produce results competitive with state-of-the-art semantic segmentation algorithms. By visualizing the intermediate layer activations of the image reconstruction module, we show that the feature map channel could correlate well with the semantic concept, which explains why joint training with the reconstruction task is helpful for the segmentation task. Motivated by our observation, we further proposed a modification to the image reconstruction task, aiming to further disentangle the object clue from the background patterns. From experiment evaluation on various datasets, we show that using reconstruction as auxiliary loss can lead to consistent improvements in various datasets and methods. The proposed method can further lead to significant improvement in object-centric segmentation tasks.

Hongrong Cheng, Miao Zhang, Javen Qinfeng Shi

As Large Language Models (LLMs) grow dramatically in size, there is an increasing trend in compressing and speeding up these models. Previous studies have highlighted the usefulness of gradients for importance scoring in neural network compressing, especially in pruning medium-size networks. However, the substantial memory requirements involved in calculating gradients with backpropagation impede the utilization of gradients in guiding LLM pruning. As a result, most pruning strategies for LLMs rely on gradient-free criteria, such as weight magnitudes or a mix of magnitudes and activations. In this paper, we devise a hybrid pruning criterion, which appropriately integrates magnitude, activation, and gradient to capitalize on feature map sensitivity for pruning LLMs. To overcome memory requirement barriers, we estimate gradients using only forward passes. Based on this, we propose a Memory-effIcieNt structured prunIng procedure for LLMs (MINI-LLM) to remove no-critical channels and multi-attention heads. Experimental results demonstrate the superior performance of MINI-LLM over existing gradient-free methods on three LLMs: LLaMA, BLOOM, and OPT across various downstream tasks (classification, multiple-choice, and generation), while MINI-LLM maintains a GPU memory footprint akin to gradient-free methods.

Hamed Damirchi, Cristian Rodriguez-Opazo, Ehsan Abbasnejad et al.

The Lottery Ticket Hypothesis (LTH) suggests that over-parameterized neural networks contain sparse subnetworks ("winning tickets") capable of matching full model performance when trained from scratch. With the growing reliance on fine-tuning large pretrained models, we investigate whether LTH extends to parameter-efficient fine-tuning (PEFT), specifically focusing on Low-Rank Adaptation (LoRA) methods. Our key finding is that LTH holds within LoRAs, revealing sparse subnetworks that can match the performance of dense adapters. In particular, we find that the effectiveness of sparse subnetworks depends more on how much sparsity is applied in each layer than on the exact weights included in the subnetwork. Building on this insight, we propose Partial-LoRA, a method that systematically identifies said subnetworks and trains sparse low-rank adapters aligned with task-relevant subspaces of the pre-trained model. Experiments across 8 vision and 12 language tasks in both single-task and multi-task settings show that Partial-LoRA reduces the number of trainable parameters by up to 87\%, while maintaining or improving accuracy. Our results not only deepen our theoretical understanding of transfer learning and the interplay between pretraining and fine-tuning but also open new avenues for developing more efficient adaptation strategies.

Jinan Zou, Maihao Guo, Yu Tian et al.

Identifying unexpected domain-shifted instances in natural language processing is crucial in real-world applications. Previous works identify the out-of-distribution (OOD) instance by leveraging a single global feature embedding to represent the sentence, which cannot characterize subtle OOD patterns well. Another major challenge current OOD methods face is learning effective low-dimensional sentence representations to identify the hard OOD instances that are semantically similar to the in-distribution (ID) data. In this paper, we propose a new unsupervised OOD detection method, namely Semantic Role Labeling Guided Out-of-distribution Detection (SRLOOD), that separates, extracts, and learns the semantic role labeling (SRL) guided fine-grained local feature representations from different arguments of a sentence and the global feature representations of the full sentence using a margin-based contrastive loss. A novel self-supervised approach is also introduced to enhance such global-local feature learning by predicting the SRL extracted role. The resulting model achieves SOTA performance on four OOD benchmarks, indicating the effectiveness of our approach. The code is publicly accessible via \url{https://github.com/cytai/SRLOOD}.

Hamed Damirchi, Ehsan Abbasnejad, Zhen Zhang et al.

Large pre-trained models have transformed machine learning, yet adapting these models effectively to exhibit precise, concept-specific behaviors remains a significant challenge. Task vectors, defined as the difference between fine-tuned and pre-trained model parameters, provide a mechanism for steering neural networks toward desired behaviors. This has given rise to large repositories dedicated to task vectors tailored for specific behaviors. The arithmetic operation of these task vectors allows for the seamless combination of desired behaviors without the need for large datasets. However, these vectors often contain overlapping concepts that can interfere with each other during arithmetic operations, leading to unpredictable outcomes. We propose a principled decomposition method that separates each task vector into two components: one capturing shared knowledge across multiple task vectors, and another isolating information unique to each specific task. By identifying invariant subspaces across projections, our approach enables more precise control over concept manipulation without unintended amplification or diminution of other behaviors. We demonstrate the effectiveness of our decomposition method across three domains: improving multi-task merging in image classification by 5% using shared components as additional task vectors, enabling clean style mixing in diffusion models without generation degradation by mixing only the unique components, and achieving 47% toxicity reduction in language models while preserving performance on general knowledge tasks by negating the toxic information isolated to the unique component. Our approach provides a new framework for understanding and controlling task vector arithmetic, addressing fundamental limitations in model editing operations.

Yuhang Liu, Zhen Zhang, Dong Gong et al.

Causal representation learning seeks to uncover latent, high-level causal representations from low-level observed data. It is particularly good at predictions under unseen distribution shifts, because these shifts can generally be interpreted as consequences of interventions. Hence leveraging {seen} distribution shifts becomes a natural strategy to help identifying causal representations, which in turn benefits predictions where distributions are previously {unseen}. Determining the types (or conditions) of such distribution shifts that do contribute to the identifiability of causal representations is critical. This work establishes a {sufficient} and {necessary} condition characterizing the types of distribution shifts for identifiability in the context of latent additive noise models. Furthermore, we present partial identifiability results when only a portion of distribution shifts meets the condition. In addition, we extend our findings to latent post-nonlinear causal models. We translate our findings into a practical algorithm, allowing for the acquisition of reliable latent causal representations. Our algorithm, guided by our underlying theory, has demonstrated outstanding performance across a diverse range of synthetic and real-world datasets. The empirical observations align closely with the theoretical findings, affirming the robustness and effectiveness of our approach.

Yuhang Liu, Dong Gong, Yichao Cai et al.

The remarkable achievements of large language models (LLMs) have led many to conclude that they exhibit a form of intelligence. This is as opposed to explanations of their capabilities based on their ability to perform relatively simple manipulations of vast volumes of data. To illuminate the distinction between these explanations, we introduce a novel generative model that generates tokens on the basis of human-interpretable concepts represented as latent discrete variables. Under mild conditions, even when the mapping from the latent space to the observed space is non-invertible, we establish an identifiability result, i.e., the representations learned by LLMs through next-token prediction can be approximately modeled as the logarithm of the posterior probabilities of these latent discrete concepts given input context, up to an invertible linear transformation. This theoretical finding not only provides evidence that LLMs capture underlying generative factors, but also provide a unified prospective for understanding of the linear representation hypothesis. Taking this a step further, our finding motivates a reliable evaluation of sparse autoencoders by treating the performance of supervised concept extractors as an upper bound. Pushing this idea even further, it inspires a structural variant that enforces dependence among latent concepts in addition to promoting sparsity. Empirically, we validate our theoretical results through evaluations on both simulation data and the Pythia, Llama, and DeepSeek model families, and demonstrate the effectiveness of our structured sparse autoencoder.

Yuhang Liu, Zhen Zhang, Dong Gong et al.

Directed acyclic graphs (DAGs) are fundamental graph structures in causal modeling, but identifying the desired DAG from observational data often requires strong assumptions that may not hold in real-world scenarios, especially for latent causal models and complex multimodal data. This raises the question of whether we can relax or bypass the DAG assumption while maintaining practical utility. In this work, we propose a novel latent partial causal model for multimodal data, featuring two latent coupled variables, connected by an undirected edge, to represent the transfer of knowledge across modalities. Under specific statistical assumptions, we establish an identifiability result, demonstrating that representations learned by multimodal contrastive learning correspond to the latent coupled variables up to a trivial transformation. This result deepens our understanding of the why multimodal contrastive learning works, highlights its potential for disentanglement, and expands the utility of pre-trained models like CLIP. Synthetic experiments confirm the robustness of our findings, even when the assumptions are partially violated. Most importantly, experiments on a pre-trained CLIP model embodies disentangled representations, enabling few-shot learning and improving domain generalization across diverse real-world datasets. Together, these contributions push the boundaries of multimodal contrastive learning, both theoretically and, crucially, in practical applications.

Yichao Cai, Yuhang Liu, Erdun Gao et al.

Multimodal representation learning, exemplified by multimodal contrastive learning (MMCL) using image-text pairs, aims to learn powerful representations by aligning cues across modalities. This approach relies on the core assumption that the exemplar image-text pairs constitute two representations of an identical concept. However, recent research has revealed that real-world datasets often exhibit cross-modal misalignment. There are two distinct viewpoints on how to address this issue: one suggests mitigating the misalignment, and the other leveraging it. We seek here to reconcile these seemingly opposing perspectives, and to provide a practical guide for practitioners. Using latent variable models we thus formalize cross-modal misalignment by introducing two specific mechanisms: Selection bias, where some semantic variables are absent in the text, and perturbation bias, where semantic variables are altered -- both leading to misalignment in data pairs. Our theoretical analysis demonstrates that, under mild assumptions, the representations learned by MMCL capture exactly the information related to the subset of the semantic variables invariant to selection and perturbation biases. This provides a unified perspective for understanding misalignment. Based on this, we further offer actionable insights into how misalignment should inform the design of real-world ML systems. We validate our theoretical findings via extensive empirical studies on both synthetic data and real image-text datasets, shedding light on the nuanced impact of cross-modal misalignment on multimodal representation learning.

Yuhao Lin, Haiming Xu, Lingqiao Liu et al.

Class-Agnostic Counting (CAC) seeks to accurately count objects in a given image with only a few reference examples. While previous methods achieving this relied on additional training, recent efforts have shown that it's possible to accomplish this without training by utilizing pre-existing foundation models, particularly the Segment Anything Model (SAM), for counting via instance-level segmentation. Although promising, current training-free methods still lag behind their training-based counterparts in terms of performance. In this research, we present a straightforward training-free solution that effectively bridges this performance gap, serving as a strong baseline. The primary contribution of our work lies in the discovery of four key technologies that can enhance performance. Specifically, we suggest employing a superpixel algorithm to generate more precise initial point prompts, utilizing an image encoder with richer semantic knowledge to replace the SAM encoder for representing candidate objects, and adopting a multiscale mechanism and a transductive prototype scheme to update the representation of reference examples. By combining these four technologies, our approach achieves significant improvements over existing training-free methods and delivers performance on par with training-based ones.

Bahram Mohammadi, Ehsan Abbasnejad, Yuankai Qi et al.

The remote embodied referring expression (REVERIE) task requires an agent to navigate through complex indoor environments and localize a remote object specified by high-level instructions, such as "bring me a spoon", without pre-exploration. Hence, an efficient navigation plan is essential for the final success. This paper proposes a novel parameter-efficient action planner using large language models (PEAP-LLM) to generate a single-step instruction at each location. The proposed model consists of two modules, LLM goal planner (LGP) and LoRA action planner (LAP). Initially, LGP extracts the goal-oriented plan from REVERIE instructions, including the target object and room. Then, LAP generates a single-step instruction with the goal-oriented plan, high-level instruction, and current visual observation as input. PEAP-LLM enables the embodied agent to interact with LAP as the path planner on the fly. A simple direct application of LLMs hardly achieves good performance. Also, existing hard-prompt-based methods are error-prone in complicated scenarios and need human intervention. To address these issues and prevent the LLM from generating hallucinations and biased information, we propose a novel two-stage method for fine-tuning the LLM, consisting of supervised fine-tuning (STF) and direct preference optimization (DPO). SFT improves the quality of generated instructions, while DPO utilizes environmental feedback. Experimental results show the superiority of our proposed model on REVERIE compared to the previous state-of-the-art.

Mingkai Tian, Guorong Li, Yuankai Qi et al.

Zero-shot video captioning requires that a model generate high-quality captions without human-annotated video-text pairs for training. State-of-the-art approaches to the problem leverage CLIP to extract visual-relevant textual prompts to guide language models in generating captions. These methods tend to focus on one key aspect of the scene and build a caption that ignores the rest of the visual input. To address this issue, and generate more accurate and complete captions, we propose a novel progressive multi-granularity textual prompting strategy for zero-shot video captioning. Our approach constructs three distinct memory banks, encompassing noun phrases, scene graphs of noun phrases, and entire sentences. Moreover, we introduce a category-aware retrieval mechanism that models the distribution of natural language surrounding the specific topics in question. Extensive experiments demonstrate the effectiveness of our method with 5.7%, 16.2%, and 3.4% improvements in terms of the main metric CIDEr on MSR-VTT, MSVD, and VATEX benchmarks compared to existing state-of-the-art.

Yuhao Lin, Lingqiao Liu, Javen Shi

Animal re-identification (ReID) has become an indispensable tool in ecological research, playing a critical role in tracking population dynamics, analyzing behavioral patterns, and assessing ecological impacts, all of which are vital for informed conservation strategies. Unlike human ReID, animal ReID faces significant challenges due to the high variability in animal poses, diverse environmental conditions, and the inability to directly apply pre-trained models to animal data, making the identification process across species more complex. This work introduces an innovative keypoint propagation mechanism, which utilizes a single annotated image and a pre-trained diffusion model to propagate keypoints across an entire dataset, significantly reducing the cost of manual annotation. Additionally, we enhance the Vision Transformer (ViT) by implementing Keypoint Positional Encoding (KPE) and Categorical Keypoint Positional Embedding (CKPE), enabling the ViT to learn more robust and semantically-aware representations. This provides more comprehensive and detailed keypoint representations, leading to more accurate and efficient re-identification. Our extensive experimental evaluations demonstrate that this approach significantly outperforms existing state-of-the-art methods across four wildlife datasets. The code will be publicly released.

Bahram Mohammadi, Yicong Hong, Yuankai Qi et al.

The vision-and-language navigation (VLN) task necessitates an agent to perceive the surroundings, follow natural language instructions, and act in photo-realistic unseen environments. Most of the existing methods employ the entire image or object features to represent navigable viewpoints. However, these representations are insufficient for proper action prediction, especially for the REVERIE task, which uses concise high-level instructions, such as ''Bring me the blue cushion in the master bedroom''. To address enhancing representation, we propose an augmented commonsense knowledge model (ACK) to leverage commonsense information as a spatio-temporal knowledge graph for improving agent navigation. Specifically, the proposed approach involves constructing a knowledge base by retrieving commonsense information from ConceptNet, followed by a refinement module to remove noisy and irrelevant knowledge. We further present ACK which consists of knowledge graph-aware cross-modal and concept aggregation modules to enhance visual representation and visual-textual data alignment by integrating visible objects, commonsense knowledge, and concept history, which includes object and knowledge temporal information. Moreover, we add a new pipeline for the commonsense-based decision-making process which leads to more accurate local action prediction. Experimental results demonstrate our proposed model noticeably outperforms the baseline and archives the state-of-the-art on the REVERIE benchmark.

Dong Gong, Qingsen Yan, Yuhang Liu et al.

Continual Learning (CL) methods aim to enable machine learning models to learn new tasks without catastrophic forgetting of those that have been previously mastered. Existing CL approaches often keep a buffer of previously-seen samples, perform knowledge distillation, or use regularization techniques towards this goal. Despite their performance, they still suffer from interference across tasks which leads to catastrophic forgetting. To ameliorate this problem, we propose to only activate and select sparse neurons for learning current and past tasks at any stage. More parameters space and model capacity can thus be reserved for the future tasks. This minimizes the interference between parameters for different tasks. To do so, we propose a Sparse neural Network for Continual Learning (SNCL), which employs variational Bayesian sparsity priors on the activations of the neurons in all layers. Full Experience Replay (FER) provides effective supervision in learning the sparse activations of the neurons in different layers. A loss-aware reservoir-sampling strategy is developed to maintain the memory buffer. The proposed method is agnostic as to the network structures and the task boundaries. Experiments on different datasets show that our approach achieves state-of-the-art performance for mitigating forgetting.

Mahdi Kazemi Moghaddam, Ehsan Abbasnejad, Qi Wu et al.

Visual navigation is often cast as a reinforcement learning (RL) problem. Current methods typically result in a suboptimal policy that learns general obstacle avoidance and search behaviours. For example, in the target-object navigation setting, the policies learnt by traditional methods often fail to complete the task, even when the target is clearly within reach from a human perspective. In order to address this issue, we propose to learn to imagine a latent representation of the successful (sub-)goal state. To do so, we have developed a module which we call Foresight Imagination (ForeSIT). ForeSIT is trained to imagine the recurrent latent representation of a future state that leads to success, e.g. either a sub-goal state that is important to reach before the target, or the goal state itself. By conditioning the policy on the generated imagination during training, our agent learns how to use this imagination to achieve its goal robustly. Our agent is able to imagine what the (sub-)goal state may look like (in the latent space) and can learn to navigate towards that state. We develop an efficient learning algorithm to train ForeSIT in an on-policy manner and integrate it into our RL objective. The integration is not trivial due to the constantly evolving state representation shared between both the imagination and the policy. We, empirically, observe that our method outperforms the state-of-the-art methods by a large margin in the commonly accepted benchmark AI2THOR environment. Our method can be readily integrated or added to other model-free RL navigation frameworks.

Zhenhua Wang, Jiajun Meng, Dongyan Guo et al.

Compared with the progress made on human activity classification, much less success has been achieved on human interaction understanding (HIU). Apart from the latter task is much more challenging, the main cause is that recent approaches learn human interactive relations via shallow graphical models, which is inadequate to model complicated human interactions. In this paper, we propose a consistency-aware graph network, which combines the representative ability of graph network and the consistency-aware reasoning to facilitate the HIU task. Our network consists of three components, a backbone CNN to extract image features, a factor graph network to learn third-order interactive relations among participants, and a consistency-aware reasoning module to enforce labeling and grouping consistencies. Our key observation is that the consistency-aware-reasoning bias for HIU can be embedded into an energy function, minimizing which delivers consistent predictions. An efficient mean-field inference algorithm is proposed, such that all modules of our network could be trained jointly in an end-to-end manner. Experimental results show that our approach achieves leading performance on three benchmarks.

Alireza Abedin, Farbod Motlagh, Qinfeng Shi et al.

Our ability to exploit low-cost wearable sensing modalities for critical human behaviour and activity monitoring applications in health and wellness is reliant on supervised learning regimes; here, deep learning paradigms have proven extremely successful in learning activity representations from annotated data. However, the costly work of gathering and annotating sensory activity datasets is labor-intensive, time consuming and not scalable to large volumes of data. While existing unsupervised remedies of deep clustering leverage network architectures and optimization objectives that are tailored for static image datasets, deep architectures to uncover cluster structures from raw sequence data captured by on-body sensors remains largely unexplored. In this paper, we develop an unsupervised end-to-end learning strategy for the fundamental problem of human activity recognition (HAR) from wearables. Through extensive experiments, including comparisons with existing methods, we show the effectiveness of our approach to jointly learn unsupervised representations for sensory data and generate cluster assignments with strong semantic correspondence to distinct human activities.

Alireza Abedin, Mahsa Ehsanpour, Qinfeng Shi et al.

Wearables are fundamental to improving our understanding of human activities, especially for an increasing number of healthcare applications from rehabilitation to fine-grained gait analysis. Although our collective know-how to solve Human Activity Recognition (HAR) problems with wearables has progressed immensely with end-to-end deep learning paradigms, several fundamental opportunities remain overlooked. We rigorously explore these new opportunities to learn enriched and highly discriminating activity representations. We propose: i) learning to exploit the latent relationships between multi-channel sensor modalities and specific activities; ii) investigating the effectiveness of data-agnostic augmentation for multi-modal sensor data streams to regularize deep HAR models; and iii) incorporating a classification loss criterion to encourage minimal intra-class representation differences whilst maximising inter-class differences to achieve more discriminative features. Our contributions achieves new state-of-the-art performance on four diverse activity recognition problem benchmarks with large margins -- with up to 6% relative margin improvement. We extensively validate the contributions from our design concepts through extensive experiments, including activity misalignment measures, ablation studies and insights shared through both quantitative and qualitative studies.

Mahsa Ehsanpour, Alireza Abedin, Fatemeh Saleh et al.

The state-of-the art solutions for human activity understanding from a video stream formulate the task as a spatio-temporal problem which requires joint localization of all individuals in the scene and classification of their actions or group activity over time. Who is interacting with whom, e.g. not everyone in a queue is interacting with each other, is often not predicted. There are scenarios where people are best to be split into sub-groups, which we call social groups, and each social group may be engaged in a different social activity. In this paper, we solve the problem of simultaneously grouping people by their social interactions, predicting their individual actions and the social activity of each social group, which we call the social task. Our main contributions are: i) we propose an end-to-end trainable framework for the social task; ii) our proposed method also sets the state-of-the-art results on two widely adopted benchmarks for the traditional group activity recognition task (assuming individuals of the scene form a single group and predicting a single group activity label for the scene); iii) we introduce new annotations on an existing group activity dataset, re-purposing it for the social task.

Qingsen Yan, Bo Wang, Dong Gong et al.

A novel coronavirus disease 2019 (COVID-19) was detected and has spread rapidly across various countries around the world since the end of the year 2019, Computed Tomography (CT) images have been used as a crucial alternative to the time-consuming RT-PCR test. However, pure manual segmentation of CT images faces a serious challenge with the increase of suspected cases, resulting in urgent requirements for accurate and automatic segmentation of COVID-19 infections. Unfortunately, since the imaging characteristics of the COVID-19 infection are diverse and similar to the backgrounds, existing medical image segmentation methods cannot achieve satisfactory performance. In this work, we try to establish a new deep convolutional neural network tailored for segmenting the chest CT images with COVID-19 infections. We firstly maintain a large and new chest CT image dataset consisting of 165,667 annotated chest CT images from 861 patients with confirmed COVID-19. Inspired by the observation that the boundary of the infected lung can be enhanced by adjusting the global intensity, in the proposed deep CNN, we introduce a feature variation block which adaptively adjusts the global properties of the features for segmenting COVID-19 infection. The proposed FV block can enhance the capability of feature representation effectively and adaptively for diverse cases. We fuse features at different scales by proposing Progressive Atrous Spatial Pyramid Pooling to handle the sophisticated infection areas with diverse appearance and shapes. We conducted experiments on the data collected in China and Germany and show that the proposed deep CNN can produce impressive performance effectively.

Mahdi Kazemi Moghaddam, Qi Wu, Ehsan Abbasnejad et al.

We humans can impeccably search for a target object, given its name only, even in an unseen environment. We argue that this ability is largely due to three main reasons: the incorporation of prior knowledge (or experience), the adaptation of it to the new environment using the observed visual cues and most importantly optimistically searching without giving up early. This is currently missing in the state-of-the-art visual navigation methods based on Reinforcement Learning (RL). In this paper, we propose to use externally learned prior knowledge of the relative object locations and integrate it into our model by constructing a neural graph. In order to efficiently incorporate the graph without increasing the state-space complexity, we propose our Graph-based Value Estimation (GVE) module. GVE provides a more accurate baseline for estimating the Advantage function in actor-critic RL algorithm. This results in reduced value estimation error and, consequently, convergence to a more optimal policy. Through empirical studies, we show that our agent, dubbed as the optimistic agent, has a more realistic estimate of the state value during a navigation episode which leads to a higher success rate. Our extensive ablation studies show the efficacy of our simple method which achieves the state-of-the-art results measured by the conventional visual navigation metrics, e.g. Success Rate (SR) and Success weighted by Path Length (SPL), in AI2THOR environment.

Patrick Dendorfer, Hamid Rezatofighi, Anton Milan et al.

Standardized benchmarks are crucial for the majority of computer vision applications. Although leaderboards and ranking tables should not be over-claimed, benchmarks often provide the most objective measure of performance and are therefore important guides for research. The benchmark for Multiple Object Tracking, MOTChallenge, was launched with the goal to establish a standardized evaluation of multiple object tracking methods. The challenge focuses on multiple people tracking, since pedestrians are well studied in the tracking community, and precise tracking and detection has high practical relevance. Since the first release, MOT15, MOT16, and MOT17 have tremendously contributed to the community by introducing a clean dataset and precise framework to benchmark multi-object trackers. In this paper, we present our MOT20benchmark, consisting of 8 new sequences depicting very crowded challenging scenes. The benchmark was presented first at the 4thBMTT MOT Challenge Workshop at the Computer Vision and Pattern Recognition Conference (CVPR) 2019, and gives to chance to evaluate state-of-the-art methods for multiple object tracking when handling extremely crowded scenarios.

Dong Gong, Wei Sun, Qinfeng Shi et al.

Most learning-based super-resolution (SR) methods aim to recover high-resolution (HR) image from a given low-resolution (LR) image via learning on LR-HR image pairs. The SR methods learned on synthetic data do not perform well in real-world, due to the domain gap between the artificially synthesized and real LR images. Some efforts are thus taken to capture real-world image pairs. The captured LR-HR image pairs usually suffer from unavoidable misalignment, which hampers the performance of end-to-end learning, however. Here, focusing on the real-world SR, we ask a different question: since misalignment is unavoidable, can we propose a method that does not need LR-HR image pairing and alignment at all and utilize real images as they are? Hence we propose a framework to learn SR from an arbitrary set of unpaired LR and HR images and see how far a step can go in such a realistic and "unsupervised" setting. To do so, we firstly train a degradation generation network to generate realistic LR images and, more importantly, to capture their distribution (i.e., learning to zoom out). Instead of assuming the domain gap has been eliminated, we minimize the discrepancy between the generated data and real data while learning a degradation adaptive SR network (i.e., learning to zoom in). The proposed unpaired method achieves state-of-the-art SR results on real-world images, even in the datasets that favor the paired-learning methods more.

Liangyi Kang, Jie Liu, Lingqiao Liu et al.

Charge prediction, determining charges for criminal cases by analyzing the textual fact descriptions, is a promising technology in legal assistant systems. In practice, the fact descriptions could exhibit a significant intra-class variation due to factors like non-normative use of language, which makes the prediction task very challenging, especially for charge classes with too few samples to cover the expression variation. In this work, we explore to use the charge definitions from criminal law to alleviate this issue. The key idea is that the expressions in a fact description should have corresponding formal terms in charge definitions, and those terms are shared across classes and could account for the diversity in the fact descriptions. Thus, we propose to create auxiliary fact representations from charge definitions to augment fact descriptions representation. The generated auxiliary representations are created through the interaction of fact description with the relevant charge definitions and terms in those definitions by integrated sentence- and word-level attention scheme. Experimental results on two datasets show that our model achieves significant improvement than baselines, especially for classes with few samples.

Yinglong Wang, Haokui Zhang, Yu Liu et al.

In recent years, deep learning based methods have made significant progress in rain-removing. However, the existing methods usually do not have good generalization ability, which leads to the fact that almost all of existing methods have a satisfied performance on removing a specific type of rain streaks, but may have a relatively poor performance on other types of rain streaks. In this paper, aiming at removing multiple types of rain streaks from single images, we propose a novel deraining framework (GRASPP-GAN), which has better generalization capacity. Specifically, a modified ResNet-18 which extracts the deep features of rainy images and a revised ASPP structure which adapts to the various shapes and sizes of rain streaks are composed together to form the backbone of our deraining network. Taking the more prominent characteristics of rain streaks in the gradient domain into consideration, a gradient loss is introduced to help to supervise our deraining training process, for which, a Sobel convolution layer is built to extract the gradient information flexibly. To further boost the performance, an adversarial learning scheme is employed for the first time to train the proposed network. Extensive experiments on both real-world and synthetic datasets demonstrate that our method outperforms the state-of-the-art deraining methods quantitatively and qualitatively. In addition, without any modifications, our proposed framework also achieves good visual performance on dehazing.

Yinglong Wang, Qinfeng Shi, Ehsan Abbasnejad et al.

Rain removal in images/videos is still an important task in computer vision field and attracting attentions of more and more people. Traditional methods always utilize some incomplete priors or filters (e.g. guided filter) to remove rain effect. Deep learning gives more probabilities to better solve this task. However, they remove rain either by evaluating background from rainy image directly or learning a rain residual first then subtracting the residual to obtain a clear background. No other models are used in deep learning based de-raining methods to remove rain and obtain other information about rainy scenes. In this paper, we utilize an extensively-used image degradation model which is derived from atmospheric scattering principles to model the formation of rainy images and try to learn the transmission, atmospheric light in rainy scenes and remove rain further. To reach this goal, we propose a robust evaluation method of global atmospheric light in a rainy scene. Instead of using the estimated atmospheric light directly to learn a network to calculate transmission, we utilize it as ground truth and design a simple but novel triangle-shaped network structure to learn atmospheric light for every rainy image, then fine-tune the network to obtain a better estimation of atmospheric light during the training of transmission network. Furthermore, more efficient ShuffleNet Units are utilized in transmission network to learn transmission map and the de-raining image is then obtained by the image degradation model. By subjective and objective comparisons, our method outperforms the selected state-of-the-art works.

Patrick Dendorfer, Hamid Rezatofighi, Anton Milan et al.

Standardized benchmarks are crucial for the majority of computer vision applications. Although leaderboards and ranking tables should not be over-claimed, benchmarks often provide the most objective measure of performance and are therefore important guides for research. The benchmark for Multiple Object Tracking, MOTChallenge, was launched with the goal to establish a standardized evaluation of multiple object tracking methods. The challenge focuses on multiple people tracking, since pedestrians are well studied in the tracking community, and precise tracking and detection has high practical relevance. Since the first release, MOT15, MOT16 and MOT17 have tremendously contributed to the community by introducing a clean dataset and precise framework to benchmark multi-object trackers. In this paper, we present our CVPR19 benchmark, consisting of 8 new sequences depicting very crowded challenging scenes. The benchmark will be presented at the 4th BMTT MOT Challenge Workshop at the Computer Vision and Pattern Recognition Conference (CVPR) 2019, and will evaluate the state-of-the-art in multiple object tracking whend handling extremely crowded scenarios.

Alireza Abedin, S. Hamid Rezatofighi, Qinfeng Shi et al.

Batteryless or so called passive wearables are providing new and innovative methods for human activity recognition (HAR), especially in healthcare applications for older people. Passive sensors are low cost, lightweight, unobtrusive and desirably disposable; attractive attributes for healthcare applications in hospitals and nursing homes. Despite the compelling propositions for sensing applications, the data streams from these sensors are characterised by high sparsity---the time intervals between sensor readings are irregular while the number of readings per unit time are often limited. In this paper, we rigorously explore the problem of learning activity recognition models from temporally sparse data. We describe how to learn directly from sparse data using a deep learning paradigm in an end-to-end manner. We demonstrate significant classification performance improvements on real-world passive sensor datasets from older people over the state-of-the-art deep learning human activity recognition models. Further, we provide insights into the model's behaviour through complementary experiments on a benchmark dataset and visualisation of the learned activity feature spaces.

Yinglong Wang, Dong Gong, Jie Yang et al.

Removing rain effects from an image is of importance for various applications such as autonomous driving, drone piloting, and photo editing. Conventional methods rely on some heuristics to handcraft various priors to remove or separate the rain effects from an image. Recent deep learning models are proposed to learn end-to-end methods to complete this task. However, they often fail to obtain satisfactory results in many realistic scenarios, especially when the observed images suffer from heavy rain. Heavy rain brings not only rain streaks but also haze-like effect caused by the accumulation of tiny raindrops. Different from the existing deep learning deraining methods that mainly focus on handling the rain streaks, we design a deep neural network by incorporating a physical raining image model. Specifically, in the proposed model, two branches are designed to handle both the rain streaks and haze-like effects. An additional submodule is jointly trained to finally refine the results, which give the model flexibility to control the strength of removing the mist. Extensive experiments on several datasets show that our method outperforms the state-of-the-art in both objective assessments and visual quality.

Mohammad Mahdi Kazemi Moghaddam, Ehsan Abbasnejad, Javen Shi

Retailers have long been searching for ways to effectively understand their customers' behaviour in order to provide a smooth and pleasant shopping experience that attracts more customers everyday and maximises their revenue, consequently. Humans can flawlessly understand others' behaviour by combining different visual cues from activity to gestures and facial expressions. Empowering the computer vision systems to do so, however, is still an open problem due to its intrinsic challenges as well as extrinsic enforced difficulties like lack of publicly available data and unique environment conditions (wild). In this work, We emphasise on detecting the first and by far the most crucial cue in behaviour analysis; that is human activity detection in computer vision. To do so, we introduce a framework for integrating human pose and object motion to both temporally detect and classify the activities in a fine-grained manner (very short and similar activities). We incorporate partial human pose and interaction with the objects in a multi-stream neural network architecture to guide the spatiotemporal attention mechanism for more efficient activity recognition. To this end, in the absence of pose supervision, we propose to use the Generative Adversarial Network (GAN) to generate exact joint locations from noisy probability heat maps. Additionally, based on the intuition that complex actions demand more than one source of information to be identified even by humans, we integrate the second stream of object motion to our network as a prior knowledge that we quantitatively show improves the recognition results. We empirically show the capability of our approach by achieving state-of-the-art results on MERL shopping dataset. We further investigate the effectiveness of this approach on a new shopping dataset that we have collected to address existing shortcomings.

Qingsen Yan, Dong Gong, Qinfeng Shi et al.

Ghosting artifacts caused by moving objects or misalignments is a key challenge in high dynamic range (HDR) imaging for dynamic scenes. Previous methods first register the input low dynamic range (LDR) images using optical flow before merging them, which are error-prone and cause ghosts in results. A very recent work tries to bypass optical flows via a deep network with skip-connections, however, which still suffers from ghosting artifacts for severe movement. To avoid the ghosting from the source, we propose a novel attention-guided end-to-end deep neural network (AHDRNet) to produce high-quality ghost-free HDR images. Unlike previous methods directly stacking the LDR images or features for merging, we use attention modules to guide the merging according to the reference image. The attention modules automatically suppress undesired components caused by misalignments and saturation and enhance desirable fine details in the non-reference images. In addition to the attention model, we use dilated residual dense block (DRDB) to make full use of the hierarchical features and increase the receptive field for hallucinating the missing details. The proposed AHDRNet is a non-flow-based method, which can also avoid the artifacts generated by optical-flow estimation error. Experiments on different datasets show that the proposed AHDRNet can achieve state-of-the-art quantitative and qualitative results.

Yong Guo, Qi Chen, Jian Chen et al.

Generating images via the generative adversarial network (GAN) has attracted much attention recently. However, most of the existing GAN-based methods can only produce low-resolution images of limited quality. Directly generating high-resolution images using GANs is nontrivial, and often produces problematic images with incomplete objects. To address this issue, we develop a novel GAN called Auto-Embedding Generative Adversarial Network (AEGAN), which simultaneously encodes the global structure features and captures the fine-grained details. In our network, we use an autoencoder to learn the intrinsic high-level structure of real images and design a novel denoiser network to provide photo-realistic details for the generated images. In the experiments, we are able to produce 512x512 images of promising quality directly from the input noise. The resultant images exhibit better perceptual photo-realism, i.e., with sharper structure and richer details, than other baselines on several datasets, including Oxford-102 Flowers, Caltech-UCSD Birds (CUB), High-Quality Large-scale CelebFaces Attributes (CelebA-HQ), Large-scale Scene Understanding (LSUN) and ImageNet.

Jie Li, Yu Liu, Dong Gong et al.

RGB images differentiate from depth images as they carry more details about the color and texture information, which can be utilized as a vital complementary to depth for boosting the performance of 3D semantic scene completion (SSC). SSC is composed of 3D shape completion (SC) and semantic scene labeling while most of the existing methods use depth as the sole input which causes the performance bottleneck. Moreover, the state-of-the-art methods employ 3D CNNs which have cumbersome networks and tremendous parameters. We introduce a light-weight Dimensional Decomposition Residual network (DDR) for 3D dense prediction tasks. The novel factorized convolution layer is effective for reducing the network parameters, and the proposed multi-scale fusion mechanism for depth and color image can improve the completion and segmentation accuracy simultaneously. Our method demonstrates excellent performance on two public datasets. Compared with the latest method SSCNet, we achieve 5.9% gains in SC-IoU and 5.7% gains in SSC-IOU, albeit with only 21% network parameters and 16.6% FLOPs employed compared with that of SSCNet.

Yu Liu, Lingqiao Liu, Hamid Rezatofighi et al.

As the post-processing step for object detection, non-maximum suppression (GreedyNMS) is widely used in most of the detectors for many years. It is efficient and accurate for sparse scenes, but suffers an inevitable trade-off between precision and recall in crowded scenes. To overcome this drawback, we propose a Pairwise-NMS to cure GreedyNMS. Specifically, a pairwise-relationship network that is based on deep learning is learned to predict if two overlapping proposal boxes contain two objects or zero/one object, which can handle multiple overlapping objects effectively. Through neatly coupling with GreedyNMS without losing efficiency, consistent improvements have been achieved in heavily occluded datasets including MOT15, TUD-Crossing and PETS. In addition, Pairwise-NMS can be integrated into any learning based detectors (Both of Faster-RCNN and DPM detectors are tested in this paper), thus building a bridge between GreedyNMS and end-to-end learning detectors.

Ehsan Abbasnejad, Qi Wu, Javen Shi et al.

One of the core challenges in Visual Dialogue problems is asking the question that will provide the most useful information towards achieving the required objective. Encouraging an agent to ask the right questions is difficult because we don't know a-priori what information the agent will need to achieve its task, and we don't have an explicit model of what it knows already. We propose a solution to this problem based on a Bayesian model of the uncertainty in the implicit model maintained by the visual dialogue agent, and in the function used to select an appropriate output. By selecting the question that minimises the predicted regret with respect to this implicit model the agent actively reduces ambiguity. The Bayesian model of uncertainty also enables a principled method for identifying when enough information has been acquired, and an action should be selected. We evaluate our approach on two goal-oriented dialogue datasets, one for visual-based collaboration task and the other for a negotiation-based task. Our uncertainty-aware information-seeking model outperforms its counterparts in these two challenging problems.