Multi-modal Queried Object Detection in the Wild

We introduce MQ-Det, an efficient architecture and pre-training strategy design to utilize both textual description with open-set generalization and visual exemplars with rich description granularity as category queries, namely, Multi-modal Queried object Detection, for real-world detection with both open-vocabulary categories and various granularity. MQ-Det incorporates vision queries into existing well-established language-queried-only detectors. A plug-and-play gated class-scalable perceiver module upon the frozen detector is proposed to augment category text with class-wise visual information. To address the learning inertia problem brought by the frozen detector, a vision conditioned masked language prediction strategy is proposed. MQ-Det's simple yet effective architecture and training strategy design is compatible with most language-queried object detectors, thus yielding versatile applications. Experimental results demonstrate that multi-modal queries largely boost open-world detection. For instance, MQ-Det significantly improves the state-of-the-art open-set detector GLIP by +7.8% zero-shot AP on the LVIS benchmark and averagely +6.3% AP on 13 few-shot downstream tasks, with merely 3% pre-training time required by GLIP. Code is available at https://github.com/YifanXu74/MQ-Det.

READ: Recurrent Adaptation of Large Transformers

Fine-tuning large-scale Transformers has led to the explosion of many AI applications across Natural Language Processing and Computer Vision tasks. However, fine-tuning all pre-trained model parameters becomes impractical as the model size and number of tasks increase. Parameter-efficient transfer learning (PETL) methods aim to address these challenges. While effective in reducing the number of trainable parameters, PETL methods still require significant energy and computational resources to fine-tune. In this paper, we introduce \textbf{RE}current \textbf{AD}aption (READ) -- a lightweight and memory-efficient fine-tuning method -- to overcome the limitations of the current PETL approaches. Specifically, READ inserts a small RNN network alongside the backbone model so that the model does not have to back-propagate through the large backbone network. Through comprehensive empirical evaluation of the GLUE benchmark, we demonstrate READ can achieve a $56\%$ reduction in the training memory consumption and an $84\%$ reduction in the GPU energy usage while retraining high model quality compared to full-tuning. Additionally, the model size of READ does not grow with the backbone model size, making it a highly scalable solution for fine-tuning large Transformers.

Integrating Multiple Sources Knowledge for Class Asymmetry Domain Adaptation Segmentation of Remote Sensing Images

In the existing unsupervised domain adaptation (UDA) methods for remote sensing images (RSIs) semantic segmentation, class symmetry is an widely followed ideal assumption, where the source and target RSIs have exactly the same class space. In practice, however, it is often very difficult to find a source RSI with exactly the same classes as the target RSI. More commonly, there are multiple source RSIs available. To this end, a novel class asymmetry RSIs domain adaptation method with multiple sources is proposed in this paper, which consists of four key components. Firstly, a multi-branch segmentation network is built to learn an expert for each source RSI. Secondly, a novel collaborative learning method with the cross-domain mixing strategy is proposed, to supplement the class information for each source while achieving the domain adaptation of each source-target pair. Thirdly, a pseudo-label generation strategy is proposed to effectively combine strengths of different experts, which can be flexibly applied to two cases where the source class union is equal to or includes the target class set. Fourthly, a multiview-enhanced knowledge integration module is developed for the high-level knowledge routing and transfer from multiple domains to target predictions.

Compressing neural network by tensor network with exponentially fewer variational parameters

Neural network (NN) designed for challenging machine learning tasks is in general a highly nonlinear mapping that contains massive variational parameters. High complexity of NN, if unbounded or unconstrained, might unpredictably cause severe issues including over-fitting, loss of generalization power, and unbearable cost of hardware. In this work, we propose a general compression scheme that significantly reduces the variational parameters of NN by encoding them to multi-layer tensor networks (TN's) that contain exponentially-fewer free parameters. Superior compression performance of our scheme is demonstrated on several widely-recognized NN's (FC-2, LeNet-5, and VGG-16) and datasets (MNIST and CIFAR-10), surpassing the state-of-the-art method based on shallow tensor networks. For instance, about 10 million parameters in the three convolutional layers of VGG-16 are compressed in TN's with just $632$ parameters, while the testing accuracy on CIFAR-10 is surprisingly improved from $81.14\%$ by the original NN to $84.36\%$ after compression. Our work suggests TN as an exceptionally efficient mathematical structure for representing the variational parameters of NN's, which superiorly exploits the compressibility than the simple multi-way arrays.

Reactive Perturbation Defocusing for Textual Adversarial Defense

Recent studies have shown that large pre-trained language models are vulnerable to adversarial attacks. Existing methods attempt to reconstruct the adversarial examples. However, these methods usually have limited performance in defense against adversarial examples, while also negatively impacting the performance on natural examples. To overcome this problem, we propose a method called Reactive Perturbation Defocusing (RPD). RPD uses an adversarial detector to identify adversarial examples and reduce false defenses on natural examples. Instead of reconstructing the adversaries, RPD injects safe perturbations into adversarial examples to distract the objective models from the malicious perturbations. Our experiments on three datasets, two objective models, and various adversarial attacks show that our proposed framework successfully repairs up to approximately 97% of correctly identified adversarial examples with only about a 2% performance decrease on natural examples. We also provide a demo of adversarial detection and repair based on our work.

DiffFacto: Controllable Part-Based 3D Point Cloud Generation with Cross Diffusion

While the community of 3D point cloud generation has witnessed a big growth in recent years, there still lacks an effective way to enable intuitive user control in the generation process, hence limiting the general utility of such methods. Since an intuitive way of decomposing a shape is through its parts, we propose to tackle the task of controllable part-based point cloud generation. We introduce DiffFacto, a novel probabilistic generative model that learns the distribution of shapes with part-level control. We propose a factorization that models independent part style and part configuration distributions and presents a novel cross-diffusion network that enables us to generate coherent and plausible shapes under our proposed factorization. Experiments show that our method is able to generate novel shapes with multiple axes of control. It achieves state-of-the-art part-level generation quality and generates plausible and coherent shapes while enabling various downstream editing applications such as shape interpolation, mixing, and transformation editing. Project website: https://difffacto.github.io/

SketchXAI: A First Look at Explainability for Human Sketches

This paper, for the very first time, introduces human sketches to the landscape of XAI (Explainable Artificial Intelligence). We argue that sketch as a ``human-centred'' data form, represents a natural interface to study explainability. We focus on cultivating sketch-specific explainability designs. This starts by identifying strokes as a unique building block that offers a degree of flexibility in object construction and manipulation impossible in photos. Following this, we design a simple explainability-friendly sketch encoder that accommodates the intrinsic properties of strokes: shape, location, and order. We then move on to define the first ever XAI task for sketch, that of stroke location inversion SLI. Just as we have heat maps for photos, and correlation matrices for text, SLI offers an explainability angle to sketch in terms of asking a network how well it can recover stroke locations of an unseen sketch. We offer qualitative results for readers to interpret as snapshots of the SLI process in the paper, and as GIFs on the project page. A minor but interesting note is that thanks to its sketch-specific design, our sketch encoder also yields the best sketch recognition accuracy to date while having the smallest number of parameters. The code is available at \url{https://sketchxai.github.io}.

RFAConv: Innovating Spatial Attention and Standard Convolutional Operation

Spatial attention has been widely used to improve the performance of convolutional neural networks. However, it has certain limitations. In this paper, we propose a new perspective on the effectiveness of spatial attention, which is that the spatial attention mechanism essentially solves the problem of convolutional kernel parameter sharing. However, the information contained in the attention map generated by spatial attention is not sufficient for large-size convolutional kernels. Therefore, we propose a novel attention mechanism called Receptive-Field Attention (RFA). Existing spatial attention, such as Convolutional Block Attention Module (CBAM) and Coordinated Attention (CA) focus only on spatial features, which does not fully address the problem of convolutional kernel parameter sharing. In contrast, RFA not only focuses on the receptive-field spatial feature but also provides effective attention weights for large-size convolutional kernels. The Receptive-Field Attention convolutional operation (RFAConv), developed by RFA, represents a new approach to replace the standard convolution operation. It offers nearly negligible increment of computational cost and parameters, while significantly improving network performance. We conducted a series of experiments on ImageNet-1k, COCO, and VOC datasets to demonstrate the superiority of our approach. Of particular importance, we believe that it is time to shift focus from spatial features to receptive-field spatial features for current spatial attention mechanisms. In this way, we can further improve network performance and achieve even better results. The code and pre-trained models for the relevant tasks can be found at https://github.com/Liuchen1997/RFAConv.

Diffusion models with location-scale noise

Diffusion Models (DMs) are powerful generative models that add Gaussian noise to the data and learn to remove it. We wanted to determine which noise distribution (Gaussian or non-Gaussian) led to better generated data in DMs. Since DMs do not work by design with non-Gaussian noise, we built a framework that allows reversing a diffusion process with non-Gaussian location-scale noise. We use that framework to show that the Gaussian distribution performs the best over a wide range of other distributions (Laplace, Uniform, t, Generalized-Gaussian).

ClusterFusion: Real-time Relative Positioning and Dense Reconstruction for UAV Cluster

As robotics technology advances, dense point cloud maps are increasingly in demand. However, dense reconstruction using a single unmanned aerial vehicle (UAV) suffers from limitations in flight speed and battery power, resulting in slow reconstruction and low coverage. Cluster UAV systems offer greater flexibility and wider coverage for map building. Existing methods of cluster UAVs face challenges with accurate relative positioning, scale drift, and high-speed dense point cloud map generation. To address these issues, we propose a cluster framework for large-scale dense reconstruction and real-time collaborative localization. The front-end of the framework is an improved visual odometry which can effectively handle large-scale scenes. Collaborative localization between UAVs is enabled through a two-stage joint optimization algorithm and a relative pose optimization algorithm, effectively achieving accurate relative positioning of UAVs and mitigating scale drift. Estimated poses are used to achieve real-time dense reconstruction and fusion of point cloud maps. To evaluate the performance of our proposed method, we conduct qualitative and quantitative experiments on real-world data. The results demonstrate that our framework can effectively suppress scale drift and generate large-scale dense point cloud maps in real-time, with the reconstruction speed increasing as more UAVs are added to the system.