Efficient Adaptation of Pre-trained Vision Transformer via Householder Transformation

A common strategy for Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViTs) involves adapting the model to downstream tasks by learning a low-rank adaptation matrix. This matrix is decomposed into a product of down-projection and up-projection matrices, with the bottleneck dimensionality being crucial for reducing the number of learnable parameters, as exemplified by prevalent methods like LoRA and Adapter. However, these low-rank strategies typically employ a fixed bottleneck dimensionality, which limits their flexibility in handling layer-wise variations. To address this limitation, we propose a novel PEFT approach inspired by Singular Value Decomposition (SVD) for representing the adaptation matrix. SVD decomposes a matrix into the product of a left unitary matrix, a diagonal matrix of scaling values, and a right unitary matrix. We utilize Householder transformations to construct orthogonal matrices that efficiently mimic the unitary matrices, requiring only a vector. The diagonal values are learned in a layer-wise manner, allowing them to flexibly capture the unique properties of each layer. This approach enables the generation of adaptation matrices with varying ranks across different layers, providing greater flexibility in adapting pre-trained models. Experiments on standard downstream vision tasks demonstrate that our method achieves promising fine-tuning performance.

Advancements in Visual Language Models for Remote Sensing: Datasets, Capabilities, and Enhancement Techniques

Recently, the remarkable success of ChatGPT has sparked a renewed wave of interest in artificial intelligence (AI), and the advancements in visual language models (VLMs) have pushed this enthusiasm to new heights. Differring from previous AI approaches that generally formulated different tasks as discriminative models, VLMs frame tasks as generative models and align language with visual information, enabling the handling of more challenging problems. The remote sensing (RS) field, a highly practical domain, has also embraced this new trend and introduced several VLM-based RS methods that have demonstrated promising performance and enormous potential. In this paper, we first review the fundamental theories related to VLM, then summarize the datasets constructed for VLMs in remote sensing and the various tasks they addressed. Finally, we categorize the improvement methods into three main parts according to the core components of VLMs and provide a detailed introduction and comparison of these methods.

TG-LLaVA: Text Guided LLaVA via Learnable Latent Embeddings

Currently, inspired by the success of vision-language models (VLMs), an increasing number of researchers are focusing on improving VLMs and have achieved promising results. However, most existing methods concentrate on optimizing the connector and enhancing the language model component, while neglecting improvements to the vision encoder itself. In contrast, we propose Text Guided LLaVA (TG-LLaVA) in this paper, which optimizes VLMs by guiding the vision encoder with text, offering a new and orthogonal optimization direction. Specifically, inspired by the purpose-driven logic inherent in human behavior, we use learnable latent embeddings as a bridge to analyze textual instruction and add the analysis results to the vision encoder as guidance, refining it. Subsequently, another set of latent embeddings extracts additional detailed text-guided information from high-resolution local patches as auxiliary information. Finally, with the guidance of text, the vision encoder can extract text-related features, similar to how humans focus on the most relevant parts of an image when considering a question. This results in generating better answers. Experiments on various datasets validate the effectiveness of the proposed method. Remarkably, without the need for additional training data, our propsoed method can bring more benefits to the baseline (LLaVA-1.5) compared with other concurrent methods. Furthermore, the proposed method consistently brings improvement in different settings.

3D-RCNet: Learning from Transformer to Build a 3D Relational ConvNet for Hyperspectral Image Classification

Recently, the Vision Transformer (ViT) model has replaced the classical Convolutional Neural Network (ConvNet) in various computer vision tasks due to its superior performance. Even in hyperspectral image (HSI) classification field, ViT-based methods also show promising potential. Nevertheless, ViT encounters notable difficulties in processing HSI data. Its self-attention mechanism, which exhibits quadratic complexity, escalates computational costs. Additionally, ViT's substantial demand for training samples does not align with the practical constraints posed by the expensive labeling of HSI data. To overcome these challenges, we propose a 3D relational ConvNet named 3D-RCNet, which inherits both strengths of ConvNet and ViT, resulting in high performance in HSI classification. We embed the self-attention mechanism of Transformer into the convolutional operation of ConvNet to design 3D relational convolutional operation and use it to build the final 3D-RCNet. The proposed 3D-RCNet maintains the high computational efficiency of ConvNet while enjoying the flexibility of ViT. Additionally, the proposed 3D relational convolutional operation is a plug-and-play operation, which can be inserted into previous ConvNet-based HSI classification methods seamlessly. Empirical evaluations on three representative benchmark HSI datasets show that the proposed model outperforms previous ConvNet-based and ViT-based HSI approaches.

OVA-DETR: Open Vocabulary Aerial Object Detection Using Image-Text Alignment and Fusion

Aerial object detection has been a hot topic for many years due to its wide application requirements. However, most existing approaches can only handle predefined categories, which limits their applicability for the open scenarios in real-world. In this paper, we extend aerial object detection to open scenarios by exploiting the relationship between image and text, and propose OVA-DETR, a high-efficiency open-vocabulary detector for aerial images. Specifically, based on the idea of image-text alignment, we propose region-text contrastive loss to replace the category regression loss in the traditional detection framework, which breaks the category limitation. Then, we propose Bidirectional Vision-Language Fusion (Bi-VLF), which includes a dual-attention fusion encoder and a multi-level text-guided Fusion Decoder. The dual-attention fusion encoder enhances the feature extraction process in the encoder part. The multi-level text-guided Fusion Decoder is designed to improve the detection ability for small objects, which frequently appear in aerial object detection scenarios. Experimental results on three widely used benchmark datasets show that our proposed method significantly improves the mAP and recall, while enjoying faster inference speed. For instance, in zero shot detection experiments on DIOR, the proposed OVA-DETR outperforms DescReg and YOLO-World by 37.4% and 33.1%, respectively, while achieving 87 FPS inference speed, which is 7.9x faster than DescReg and 3x faster than YOLO-world. The code is available at https://github.com/GT-Wei/OVA-DETR.

Bridging Sensor Gaps via Single-Direction Tuning for Hyperspectral Image Classification

Recently, some researchers started exploring the use of ViTs in tackling HSI classification and achieved remarkable results. However, the training of ViT models requires a considerable number of training samples, while hyperspectral data, due to its high annotation costs, typically has a relatively small number of training samples. This contradiction has not been effectively addressed. In this paper, aiming to solve this problem, we propose the single-direction tuning (SDT) strategy, which serves as a bridge, allowing us to leverage existing labeled HSI datasets even RGB datasets to enhance the performance on new HSI datasets with limited samples. The proposed SDT inherits the idea of prompt tuning, aiming to reuse pre-trained models with minimal modifications for adaptation to new tasks. But unlike prompt tuning, SDT is custom-designed to accommodate the characteristics of HSIs. The proposed SDT utilizes a parallel architecture, an asynchronous cold-hot gradient update strategy, and unidirectional interaction. It aims to fully harness the potent representation learning capabilities derived from training on heterologous, even cross-modal datasets. In addition, we also introduce a novel Triplet-structured transformer (Tri-Former), where spectral attention and spatial attention modules are merged in parallel to construct the token mixing component for reducing computation cost and a 3D convolution-based channel mixer module is integrated to enhance stability and keep structure information. Comparison experiments conducted on three representative HSI datasets captured by different sensors demonstrate the proposed Tri-Former achieves better performance compared to several state-of-the-art methods. Homologous, heterologous and cross-modal tuning experiments verified the effectiveness of the proposed SDT.

NAR-Former V2: Rethinking Transformer for Universal Neural Network Representation Learning

As more deep learning models are being applied in real-world applications, there is a growing need for modeling and learning the representations of neural networks themselves. An efficient representation can be used to predict target attributes of networks without the need for actual training and deployment procedures, facilitating efficient network deployment and design. Recently, inspired by the success of Transformer, some Transformer-based representation learning frameworks have been proposed and achieved promising performance in handling cell-structured models. However, graph neural network (GNN) based approaches still dominate the field of learning representation for the entire network. In this paper, we revisit Transformer and compare it with GNN to analyse their different architecture characteristics. We then propose a modified Transformer-based universal neural network representation learning model NAR-Former V2. It can learn efficient representations from both cell-structured networks and entire networks. Specifically, we first take the network as a graph and design a straightforward tokenizer to encode the network into a sequence. Then, we incorporate the inductive representation learning capability of GNN into Transformer, enabling Transformer to generalize better when encountering unseen architecture. Additionally, we introduce a series of simple yet effective modifications to enhance the ability of the Transformer in learning representation from graph structures. Our proposed method surpasses the GNN-based method NNLP by a significant margin in latency estimation on the NNLQP dataset. Furthermore, regarding accuracy prediction on the NASBench101 and NASBench201 datasets, our method achieves highly comparable performance to other state-of-the-art methods.

Teacher Agent: A Non-Knowledge Distillation Method for Rehearsal-based Video Incremental Learning

With the rise in popularity of video-based social media, new categories of videos are constantly being generated, creating an urgent need for robust incremental learning techniques for video understanding. One of the biggest challenges in this task is catastrophic forgetting, where the network tends to forget previously learned data while learning new categories. To overcome this issue, knowledge distillation is a widely used technique for rehearsal-based video incremental learning that involves transferring important information on similarities among different categories to enhance the student model. Therefore, it is preferable to have a strong teacher model to guide the students. However, the limited performance of the network itself and the occurrence of catastrophic forgetting can result in the teacher network making inaccurate predictions for some memory exemplars, ultimately limiting the student network's performance. Based on these observations, we propose a teacher agent capable of generating stable and accurate soft labels to replace the output of the teacher model. This method circumvents the problem of knowledge misleading caused by inaccurate predictions of the teacher model and avoids the computational overhead of loading the teacher model for knowledge distillation. Extensive experiments demonstrate the advantages of our method, yielding significant performance improvements while utilizing only half the resolution of video clips in the incremental phases as input compared to recent state-of-the-art methods. Moreover, our method surpasses the performance of joint training when employing four times the number of samples in episodic memory.

NAR-Former: Neural Architecture Representation Learning towards Holistic Attributes Prediction

With the wide and deep adoption of deep learning models in real applications, there is an increasing need to model and learn the representations of the neural networks themselves. These models can be used to estimate attributes of different neural network architectures such as the accuracy and latency, without running the actual training or inference tasks. In this paper, we propose a neural architecture representation model that can be used to estimate these attributes holistically. Specifically, we first propose a simple and effective tokenizer to encode both the operation and topology information of a neural network into a single sequence. Then, we design a multi-stage fusion transformer to build a compact vector representation from the converted sequence. For efficient model training, we further propose an information flow consistency augmentation and correspondingly design an architecture consistency loss, which brings more benefits with less augmentation samples compared with previous random augmentation strategies. Experiment results on NAS-Bench-101, NAS-Bench-201, DARTS search space and NNLQP show that our proposed framework can be used to predict the aforementioned latency and accuracy attributes of both cell architectures and whole deep neural networks, and achieves promising performance.

CabViT: Cross Attention among Blocks for Vision Transformer

Since the vision transformer (ViT) has achieved impressive performance in image classification, an increasing number of researchers pay their attentions to designing more efficient vision transformer models. A general research line is reducing computational cost of self attention modules by adopting sparse attention or using local attention windows. In contrast, we propose to design high performance transformer based architectures by densifying the attention pattern. Specifically, we propose cross attention among blocks of ViT (CabViT), which uses tokens from previous blocks in the same stage as extra input to the multi-head attention of transformers. The proposed CabViT enhances the interactions of tokens across blocks with potentially different semantics, and encourages more information flows to the lower levels, which together improves model performance and model convergence with limited extra cost. Based on the proposed CabViT, we design a series of CabViT models which achieve the best trade-off between model size, computational cost and accuracy. For instance without the need of knowledge distillation to strength the training, CabViT achieves 83.0% top-1 accuracy on Imagenet with only 16.3 million parameters and about 3.9G FLOPs, saving almost half parameters and 13% computational cost while gaining 0.9% higher accuracy compared with ConvNext, use 52% of parameters but gaining 0.6% accuracy compared with distilled EfficientFormer