arXiv subjects

Han Hu

Publications and source records attributed to Han Hu.

At least 163 records · Page 9Linked to original sources

arXiv2023

Rank-DETR for High Quality Object Detection

Modern detection transformers (DETRs) use a set of object queries to predict a list of bounding boxes, sort them by their classification confidence scores, and select the top-ranked predictions as the final detection results for the given input image. A highly performant object detector requires accurate ranking for the bounding box predictions. For DETR-based detectors, the top-ranked bounding boxes suffer from less accurate localization quality due to the misalignment between classification scores and localization accuracy, thus impeding the construction of high-quality detectors. In this work, we introduce a simple and highly performant DETR-based object detector by proposing a series of rank-oriented designs, combinedly called Rank-DETR. Our key contributions include: (i) a rank-oriented architecture design that can prompt positive predictions and suppress the negative ones to ensure lower false positive rates, as well as (ii) a rank-oriented loss function and matching cost design that prioritizes predictions of more accurate localization accuracy during ranking to boost the AP under high IoU thresholds. We apply our method to improve the recent SOTA methods (e.g., H-DETR and DINO-DETR) and report strong COCO object detection results when using different backbones such as ResNet-$50$, Swin-T, and Swin-L, demonstrating the effectiveness of our approach. Code is available at \url{https://github.com/LeapLabTHU/Rank-DETR}.

cs.CV↗

arXiv2023

One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge Distillation

Knowledge distillation~(KD) has proven to be a highly effective approach for enhancing model performance through a teacher-student training scheme. However, most existing distillation methods are designed under the assumption that the teacher and student models belong to the same model family, particularly the hint-based approaches. By using centered kernel alignment (CKA) to compare the learned features between heterogeneous teacher and student models, we observe significant feature divergence. This divergence illustrates the ineffectiveness of previous hint-based methods in cross-architecture distillation. To tackle the challenge in distilling heterogeneous models, we propose a simple yet effective one-for-all KD framework called OFA-KD, which significantly improves the distillation performance between heterogeneous architectures. Specifically, we project intermediate features into an aligned latent space such as the logits space, where architecture-specific information is discarded. Additionally, we introduce an adaptive target enhancement scheme to prevent the student from being disturbed by irrelevant information. Extensive experiments with various architectures, including CNN, Transformer, and MLP, demonstrate the superiority of our OFA-KD framework in enabling distillation between heterogeneous architectures. Specifically, when equipped with our OFA-KD, the student models achieve notable performance improvements, with a maximum gain of 8.0% on the CIFAR-100 dataset and 0.7% on the ImageNet-1K dataset. PyTorch code and checkpoints can be found at https://github.com/Hao840/OFAKD.

cs.CV↗

arXiv2023

Boosting output performance of contact-separation mode triboelectric nanogenerators by adopting discontinuity and fringing effect: experiment and modelling studies

Triboelectric nanogenerators (TENGs) are promising self-powering supplies for a diverse range of intelligent sensing and monitoring devices, especially due to their capability of harvesting electric energy from low frequency and small-scale mechanical motions. Inspired by the fact that contact-separation mode TENGs with small contact areas harvest high electrical outputs due to fringing effect, this study employed discontinuity on the dielectric side of contact-separation mode TENGs to promote fringing electric fields for the enhancement of electrical outputs. The results reveal that the TENGs with more discontinuities show higher overall electric performance. Compared to pristine TENGs, the TENGs with cross discontinuities increased the surface charge by 50% and the power density by 114%. However, one should avoid generating discontinuities on tribonegative side of TENGs using metal blade within a positive-ion atmosphere due to the neutralization through electrically conductive metal blade. The computational simulation validated that the TENGs with discontinuities obtained higher electrical outputs, and further investigated the effect of discontinuity gap size and array distance on TENGs performance. This study has provided a promising method for the future design of TENGs using discontinuous structures.

physics.app-ph↗

arXiv2023

Attentive Mask CLIP

Image token removal is an efficient augmentation strategy for reducing the cost of computing image features. However, this efficient augmentation strategy has been found to adversely affect the accuracy of CLIP-based training. We hypothesize that removing a large portion of image tokens may improperly discard the semantic content associated with a given text description, thus constituting an incorrect pairing target in CLIP training. To address this issue, we propose an attentive token removal approach for CLIP training, which retains tokens with a high semantic correlation to the text description. The correlation scores are computed in an online fashion using the EMA version of the visual encoder. Our experiments show that the proposed attentive masking approach performs better than the previous method of random token removal for CLIP training. The approach also makes it efficient to apply multiple augmentation views to the image, as well as introducing instance contrastive learning tasks between these views into the CLIP framework. Compared to other CLIP improvements that combine different pre-training targets such as SLIP and MaskCLIP, our method is not only more effective, but also much more efficient. Specifically, using ViT-B and YFCC-15M dataset, our approach achieves $43.9\%$ top-1 accuracy on ImageNet-1K zero-shot classification, as well as $62.7/42.1$ and $38.0/23.2$ I2T/T2I retrieval accuracy on Flickr30K and MS COCO, which are $+1.1\%$, $+5.5/+0.9$, and $+4.4/+1.3$ higher than the SLIP method, while being $2.30\times$ faster. An efficient version of our approach running $1.16\times$ faster than the plain CLIP model achieves significant gains of $+5.3\%$, $+11.3/+8.0$, and $+9.5/+4.9$ on these benchmarks.

cs.CV↗

arXiv2023

Domain-knowledge Inspired Pseudo Supervision (DIPS) for Unsupervised Image-to-Image Translation Models to Support Cross-Domain Classification

The ability to classify images is dependent on having access to large labeled datasets and testing on data from the same domain that the model can train on. Classification becomes more challenging when dealing with new data from a different domain, where gathering and especially labeling a larger image dataset for retraining a classification model requires a labor-intensive human effort. Cross-domain classification frameworks were developed to handle this data domain shift problem by utilizing unsupervised image-to-image translation models to translate an input image from the unlabeled domain to the labeled domain. The problem with these unsupervised models lies in their unsupervised nature. For lack of annotations, it is not possible to use the traditional supervised metrics to evaluate these translation models to pick the best-saved checkpoint model. This paper introduces a new method called Domain-knowledge Inspired Pseudo Supervision (DIPS) which utilizes domain-informed Gaussian Mixture Models to generate pseudo annotations to enable the use of traditional supervised metrics. This method was designed specifically to support cross-domain classification applications contrary to other typically used metrics such as the FID which were designed to evaluate the model in terms of the quality of the generated image from a human-eye perspective. DIPS proves its effectiveness by outperforming various GAN evaluation metrics, including FID, when selecting the optimal saved checkpoint model. It is also evaluated against truly supervised metrics. Furthermore, DIPS showcases its robustness and interpretability by demonstrating a strong correlation with truly supervised metrics, highlighting its superiority over existing state-of-the-art alternatives. The code and data to replicate the results can be found on the official Github repository: https://github.com/Hindawi91/DIPS

cs.CV↗

arXiv2023

Deep Model Fusion: A Survey

Deep model fusion/merging is an emerging technique that merges the parameters or predictions of multiple deep learning models into a single one. It combines the abilities of different models to make up for the biases and errors of a single model to achieve better performance. However, deep model fusion on large-scale deep learning models (e.g., LLMs and foundation models) faces several challenges, including high computational cost, high-dimensional parameter space, interference between different heterogeneous models, etc. Although model fusion has attracted widespread attention due to its potential to solve complex real-world tasks, there is still a lack of complete and detailed survey research on this technique. Accordingly, in order to understand the model fusion method better and promote its development, we present a comprehensive survey to summarize the recent progress. Specifically, we categorize existing deep model fusion methods as four-fold: (1) "Mode connectivity", which connects the solutions in weight space via a path of non-increasing loss, in order to obtain better initialization for model fusion; (2) "Alignment" matches units between neural networks to create better conditions for fusion; (3) "Weight average", a classical model fusion method, averages the weights of multiple models to obtain more accurate results closer to the optimal solution; (4) "Ensemble learning" combines the outputs of diverse models, which is a foundational technique for improving the accuracy and robustness of the final model. In addition, we analyze the challenges faced by deep model fusion and propose possible research directions for model fusion in the future. Our review is helpful in deeply understanding the correlation between different model fusion methods and practical application methods, which can enlighten the research in the field of deep model fusion.

cs.LG↗

arXiv2023

TinyCLIP: CLIP Distillation via Affinity Mimicking and Weight Inheritance

In this paper, we propose a novel cross-modal distillation method, called TinyCLIP, for large-scale language-image pre-trained models. The method introduces two core techniques: affinity mimicking and weight inheritance. Affinity mimicking explores the interaction between modalities during distillation, enabling student models to mimic teachers' behavior of learning cross-modal feature alignment in a visual-linguistic affinity space. Weight inheritance transmits the pre-trained weights from the teacher models to their student counterparts to improve distillation efficiency. Moreover, we extend the method into a multi-stage progressive distillation to mitigate the loss of informative weights during extreme compression. Comprehensive experiments demonstrate the efficacy of TinyCLIP, showing that it can reduce the size of the pre-trained CLIP ViT-B/32 by 50%, while maintaining comparable zero-shot performance. While aiming for comparable performance, distillation with weight inheritance can speed up the training by 1.4 - 7.8 $\times$ compared to training from scratch. Moreover, our TinyCLIP ViT-8M/16, trained on YFCC-15M, achieves an impressive zero-shot top-1 accuracy of 41.1% on ImageNet, surpassing the original CLIP ViT-B/16 by 3.5% while utilizing only 8.9% parameters. Finally, we demonstrate the good transferability of TinyCLIP in various downstream tasks. Code and models will be open-sourced at https://aka.ms/tinyclip.

cs.CV↗

arXiv2023

Exploring Non-additive Randomness on ViT against Query-Based Black-Box Attacks

Deep Neural Networks can be easily fooled by small and imperceptible perturbations. The query-based black-box attack (QBBA) is able to create the perturbations using model output probabilities of image queries requiring no access to the underlying models. QBBA poses realistic threats to real-world applications. Recently, various types of robustness have been explored to defend against QBBA. In this work, we first taxonomize the stochastic defense strategies against QBBA. Following our taxonomy, we propose to explore non-additive randomness in models to defend against QBBA. Specifically, we focus on underexplored Vision Transformers based on their flexible architectures. Extensive experiments show that the proposed defense approach achieves effective defense, without much sacrifice in performance.

cs.CV↗

arXiv2023

DeViT: Decomposing Vision Transformers for Collaborative Inference in Edge Devices

Recent years have witnessed the great success of vision transformer (ViT), which has achieved state-of-the-art performance on multiple computer vision benchmarks. However, ViT models suffer from vast amounts of parameters and high computation cost, leading to difficult deployment on resource-constrained edge devices. Existing solutions mostly compress ViT models to a compact model but still cannot achieve real-time inference. To tackle this issue, we propose to explore the divisibility of transformer structure, and decompose the large ViT into multiple small models for collaborative inference at edge devices. Our objective is to achieve fast and energy-efficient collaborative inference while maintaining comparable accuracy compared with large ViTs. To this end, we first propose a collaborative inference framework termed DeViT to facilitate edge deployment by decomposing large ViTs. Subsequently, we design a decomposition-and-ensemble algorithm based on knowledge distillation, termed DEKD, to fuse multiple small decomposed models while dramatically reducing communication overheads, and handle heterogeneous models by developing a feature matching module to promote the imitations of decomposed models from the large ViT. Extensive experiments for three representative ViT backbones on four widely-used datasets demonstrate our method achieves efficient collaborative inference for ViTs and outperforms existing lightweight ViTs, striking a good trade-off between efficiency and accuracy. For example, our DeViTs improves end-to-end latency by 2.89$\times$ with only 1.65% accuracy sacrifice using CIFAR-100 compared to the large ViT, ViT-L/16, on the GPU server. DeDeiTs surpasses the recent efficient ViT, MobileViT-S, by 3.54% in accuracy on ImageNet-1K, while running 1.72$\times$ faster and requiring 55.28% lower energy consumption on the edge device.

cs.CV↗

arXiv2023

InstructDiffusion: A Generalist Modeling Interface for Vision Tasks

We present InstructDiffusion, a unifying and generic framework for aligning computer vision tasks with human instructions. Unlike existing approaches that integrate prior knowledge and pre-define the output space (e.g., categories and coordinates) for each vision task, we cast diverse vision tasks into a human-intuitive image-manipulating process whose output space is a flexible and interactive pixel space. Concretely, the model is built upon the diffusion process and is trained to predict pixels according to user instructions, such as encircling the man's left shoulder in red or applying a blue mask to the left car. InstructDiffusion could handle a variety of vision tasks, including understanding tasks (such as segmentation and keypoint detection) and generative tasks (such as editing and enhancement). It even exhibits the ability to handle unseen tasks and outperforms prior methods on novel datasets. This represents a significant step towards a generalist modeling interface for vision tasks, advancing artificial general intelligence in the field of computer vision.

cs.CV↗

arXiv2023

Mask-Attention-Free Transformer for 3D Instance Segmentation

Recently, transformer-based methods have dominated 3D instance segmentation, where mask attention is commonly involved. Specifically, object queries are guided by the initial instance masks in the first cross-attention, and then iteratively refine themselves in a similar manner. However, we observe that the mask-attention pipeline usually leads to slow convergence due to low-recall initial instance masks. Therefore, we abandon the mask attention design and resort to an auxiliary center regression task instead. Through center regression, we effectively overcome the low-recall issue and perform cross-attention by imposing positional prior. To reach this goal, we develop a series of position-aware designs. First, we learn a spatial distribution of 3D locations as the initial position queries. They spread over the 3D space densely, and thus can easily capture the objects in a scene with a high recall. Moreover, we present relative position encoding for the cross-attention and iterative refinement for more accurate position queries. Experiments show that our approach converges 4x faster than existing work, sets a new state of the art on ScanNetv2 3D instance segmentation benchmark, and also demonstrates superior performance across various datasets. Code and models are available at https://github.com/dvlab-research/Mask-Attention-Free-Transformer.

cs.CV↗

arXiv2023

Over-the-Air Computation Aided Federated Learning with the Aggregation of Normalized Gradient

Over-the-air computation is a communication-efficient solution for federated learning (FL). In such a system, iterative procedure is performed: Local gradient of private loss function is updated, amplified and then transmitted by every mobile device; the server receives the aggregated gradient all-at-once, generates and then broadcasts updated model parameters to every mobile device. In terms of amplification factor selection, most related works suppose the local gradient's maximal norm always happens although it actually fluctuates over iterations, which may degrade convergence performance. To circumvent this problem, we propose to turn local gradient to be normalized one before amplifying it. Under our proposed method, when the loss function is smooth, we prove our proposed method can converge to stationary point at sub-linear rate. In case of smooth and strongly convex loss function, we prove our proposed method can achieve minimal training loss at linear rate with any small positive tolerance. Moreover, a tradeoff between convergence rate and the tolerance is discovered. To speedup convergence, problems optimizing system parameters are also formulated for above two cases. Although being non-convex, optimal solution with polynomial complexity of the formulated problems are derived. Experimental results show our proposed method can outperform benchmark methods on convergence performance.

cs.LG↗

arXiv2023

PartSeg: Few-shot Part Segmentation via Part-aware Prompt Learning

In this work, we address the task of few-shot part segmentation, which aims to segment the different parts of an unseen object using very few labeled examples. It is found that leveraging the textual space of a powerful pre-trained image-language model (such as CLIP) can be beneficial in learning visual features. Therefore, we develop a novel method termed PartSeg for few-shot part segmentation based on multimodal learning. Specifically, we design a part-aware prompt learning method to generate part-specific prompts that enable the CLIP model to better understand the concept of ``part'' and fully utilize its textual space. Furthermore, since the concept of the same part under different object categories is general, we establish relationships between these parts during the prompt learning process. We conduct extensive experiments on the PartImageNet and Pascal$\_$Part datasets, and the experimental results demonstrated that our proposed method achieves state-of-the-art performance.

cs.CV↗

arXiv2023

Federated Learning Robust to Byzantine Attacks: Achieving Zero Optimality Gap

In this paper, we propose a robust aggregation method for federated learning (FL) that can effectively tackle malicious Byzantine attacks. At each user, model parameter is firstly updated by multiple steps, which is adjustable over iterations, and then pushed to the aggregation center directly. This decreases the number of interactions between the aggregation center and users, allows each user to set training parameter in a flexible way, and reduces computation burden compared with existing works that need to combine multiple historical model parameters. At the aggregation center, geometric median is leveraged to combine the received model parameters from each user. Rigorous proof shows that zero optimality gap is achieved by our proposed method with linear convergence, as long as the fraction of Byzantine attackers is below half. Numerical results verify the effectiveness of our proposed method.

cs.LG↗

arXiv2023

SimDA: Simple Diffusion Adapter for Efficient Video Generation

The recent wave of AI-generated content has witnessed the great development and success of Text-to-Image (T2I) technologies. By contrast, Text-to-Video (T2V) still falls short of expectations though attracting increasing interests. Existing works either train from scratch or adapt large T2I model to videos, both of which are computation and resource expensive. In this work, we propose a Simple Diffusion Adapter (SimDA) that fine-tunes only 24M out of 1.1B parameters of a strong T2I model, adapting it to video generation in a parameter-efficient way. In particular, we turn the T2I model for T2V by designing light-weight spatial and temporal adapters for transfer learning. Besides, we change the original spatial attention to the proposed Latent-Shift Attention (LSA) for temporal consistency. With similar model architecture, we further train a video super-resolution model to generate high-definition (1024x1024) videos. In addition to T2V generation in the wild, SimDA could also be utilized in one-shot video editing with only 2 minutes tuning. Doing so, our method could minimize the training effort with extremely few tunable parameters for model adaptation.

cs.CV↗

arXiv2023

Joint Power Control and Data Size Selection for Over-the-Air Computation Aided Federated Learning

Federated learning (FL) has emerged as an appealing machine learning approach to deal with massive raw data generated at multiple mobile devices, {which needs to aggregate the training model parameter of every mobile device at one base station (BS) iteratively}. For parameter aggregating in FL, over-the-air computation is a spectrum-efficient solution, which allows all mobile devices to transmit their parameter-mapped signals concurrently to a BS. Due to heterogeneous channel fading and noise, there exists difference between the BS's received signal and its desired signal, measured as the mean-squared error (MSE). To minimize the MSE, we propose to jointly optimize the signal amplification factors at the BS and the mobile devices as well as the data size (the number of data samples involved in local training) at every mobile device. The formulated problem is challenging to solve due to its non-convexity. To find the optimal solution, with some simplification on cost function and variable replacement, which still preserves equivalence, we transform the changed problem to be a bi-level problem equivalently. For the lower-level problem, optimal solution is found by enumerating every candidate solution from the Karush-Kuhn-Tucker (KKT) condition. For the upper-level problem, the optimal solution is found by exploring its piecewise convexity. Numerical results show that our proposed method can greatly reduce the MSE and can help to improve the training performance of FL compared with benchmark methods.

cs.LG↗

arXiv2023

Implicit Temporal Modeling with Learnable Alignment for Video Recognition

Contrastive language-image pretraining (CLIP) has demonstrated remarkable success in various image tasks. However, how to extend CLIP with effective temporal modeling is still an open and crucial problem. Existing factorized or joint spatial-temporal modeling trades off between the efficiency and performance. While modeling temporal information within straight through tube is widely adopted in literature, we find that simple frame alignment already provides enough essence without temporal attention. To this end, in this paper, we proposed a novel Implicit Learnable Alignment (ILA) method, which minimizes the temporal modeling effort while achieving incredibly high performance. Specifically, for a frame pair, an interactive point is predicted in each frame, serving as a mutual information rich region. By enhancing the features around the interactive point, two frames are implicitly aligned. The aligned features are then pooled into a single token, which is leveraged in the subsequent spatial self-attention. Our method allows eliminating the costly or insufficient temporal self-attention in video. Extensive experiments on benchmarks demonstrate the superiority and generality of our module. Particularly, the proposed ILA achieves a top-1 accuracy of 88.7% on Kinetics-400 with much fewer FLOPs compared with Swin-L and ViViT-H. Code is released at https://github.com/Francis-Rings/ILA .

cs.CV↗

arXiv2023

Rethinking the Localization in Weakly Supervised Object Localization

Weakly supervised object localization (WSOL) is one of the most popular and challenging tasks in computer vision. This task is to localize the objects in the images given only the image-level supervision. Recently, dividing WSOL into two parts (class-agnostic object localization and object classification) has become the state-of-the-art pipeline for this task. However, existing solutions under this pipeline usually suffer from the following drawbacks: 1) they are not flexible since they can only localize one object for each image due to the adopted single-class regression (SCR) for localization; 2) the generated pseudo bounding boxes may be noisy, but the negative impact of such noise is not well addressed. To remedy these drawbacks, we first propose to replace SCR with a binary-class detector (BCD) for localizing multiple objects, where the detector is trained by discriminating the foreground and background. Then we design a weighted entropy (WE) loss using the unlabeled data to reduce the negative impact of noisy bounding boxes. Extensive experiments on the popular CUB-200-2011 and ImageNet-1K datasets demonstrate the effectiveness of our method.

cs.CV↗