An Memory-Efficient Framework for Deformable Transformer with Neural Architecture Search

• URL: http://arxiv.org/abs/2507.11549v1
• Date: Sun, 13 Jul 2025 06:42:00 GMT
• Title: An Memory-Efficient Framework for Deformable Transformer with Neural Architecture Search
• Abstract summary: Deformable Attention Transformers ( DAT) have shown remarkable performance in computer vision tasks.<n>Existing acceleration methods either incur high hardware overhead or compromise model accuracy.<n>This paper proposes a hardware-friendly optimization framework for DAT.
• Score: 4.533437219543447
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deformable Attention Transformers (DAT) have shown remarkable performance in computer vision tasks by adaptively focusing on informative image regions. However, their data-dependent sampling mechanism introduces irregular memory access patterns, posing significant challenges for efficient hardware deployment. Existing acceleration methods either incur high hardware overhead or compromise model accuracy. To address these issues, this paper proposes a hardware-friendly optimization framework for DAT. First, a neural architecture search (NAS)-based method with a new slicing strategy is proposed to automatically divide the input feature into uniform patches during the inference process, avoiding memory conflicts without modifying model architecture. The method explores the optimal slice configuration by jointly optimizing hardware cost and inference accuracy. Secondly, an FPGA-based verification system is designed to test the performance of this framework on edge-side hardware. Algorithm experiments on the ImageNet-1K dataset demonstrate that our hardware-friendly framework can maintain have only 0.2% accuracy drop compared to the baseline DAT. Hardware experiments on Xilinx FPGA show the proposed method reduces DRAM access times to 18% compared with existing DAT acceleration methods.

Related papers

• Hardware-aware Graph Neural Networks prunning for embedded event-based vision [6.945765172815977]
  Event-based cameras are gaining popularity as the sensor of choice for mobile robotics.<n>These applications require efficient real-time data processing with low latency and power consumption.<n>One strategy to meet these stringent requirements is hardware acceleration of efficient algorithms.
  arXiv  Detail & Related papers  (2026-07-07T19:09:03Z)
• Event-based Visual Deformation Measurement [76.25283405575108]
  Visual Deformation Measurement aims to recover dense deformation fields by tracking surface motion from camera observations.<n>Traditional image-based methods rely on minimal inter-frame motion to constrain the correspondence search space.<n>We propose an event-frame fusion framework that exploits events for temporally dense motion cues and frames for spatially dense precise estimation.
  arXiv  Detail & Related papers  (2026-02-16T01:04:48Z)
• Rethinking Autoregressive Models for Lossless Image Compression via Hierarchical Parallelism and Progressive Adaptation [75.58269386927076]
  Autoregressive (AR) models are often dismissed as impractical due to prohibitive computational cost.<n>This work re-thinks this paradigm, introducing a framework built on hierarchical parallelism and progressive adaptation.<n> Experiments on diverse datasets (natural, satellite, medical) validate that our method achieves new state-of-the-art compression.
  arXiv  Detail & Related papers  (2025-11-14T06:27:58Z)
• Federated Learning of Low-Rank One-Shot Image Detection Models in Edge Devices with Scalable Accuracy and Compute Complexity [5.820612543019548]
  LoRa-FL is designed for training low-rank one-shot image detection models deployed on edge devices.<n>By incorporating low-rank adaptation techniques into one-shot detection architectures, our method significantly reduces both computational and communication overhead.
  arXiv  Detail & Related papers  (2025-04-23T08:40:44Z)
• Efficient Multi-Instance Generation with Janus-Pro-Dirven Prompt Parsing [53.295515505026096]
  Janus-Pro-driven Prompt Parsing is a prompt- parsing module that bridges text understanding and layout generation.<n>MIGLoRA is a parameter-efficient plug-in integrating Low-Rank Adaptation into UNet (SD1.5) and DiT (SD3) backbones.<n>The proposed method achieves state-of-the-art performance on COCO and LVIS benchmarks while maintaining parameter efficiency.
  arXiv  Detail & Related papers  (2025-03-27T00:59:14Z)
• XR-VIO: High-precision Visual Inertial Odometry with Fast Initialization for XR Applications [34.2082611110639]
  This paper presents a novel approach to Visual Inertial Odometry (VIO) focusing on the initialization and feature matching modules.<n>Existing methods for gyroscopes often suffer from poor stability in visual Structure from Motion (SfM) or in solving a huge number of parameters simultaneously.<n>By tightly coupling measurements, we enhance the robustness and accuracy of visual SfM.<n>In terms of feature matching, we introduce a hybrid method that combines optical flow and descriptor-based matching.
  arXiv  Detail & Related papers  (2025-02-03T12:17:51Z)
• Task-Oriented Real-time Visual Inference for IoVT Systems: A Co-design Framework of Neural Networks and Edge Deployment [61.20689382879937]
  Task-oriented edge computing addresses this by shifting data analysis to the edge. Existing methods struggle to balance high model performance with low resource consumption. We propose a novel co-design framework to optimize neural network architecture.
  arXiv  Detail & Related papers  (2024-10-29T19:02:54Z)
• Quasar-ViT: Hardware-Oriented Quantization-Aware Architecture Search for Vision Transformers [56.37495946212932]
  Vision transformers (ViTs) have demonstrated their superior accuracy for computer vision tasks compared to convolutional neural networks (CNNs) This work proposes Quasar-ViT, a hardware-oriented quantization-aware architecture search framework for ViTs.
  arXiv  Detail & Related papers  (2024-07-25T16:35:46Z)
• Regularized Adaptive Momentum Dual Averaging with an Efficient Inexact Subproblem Solver for Training Structured Neural Network [9.48424754175943]
  We propose a Regularized Adaptive Momentum Dual Averaging (RAMDA) for training structured neural networks.<n>We show that RAMDA attains the ideal structure induced by the regularizer at the stationary point of convergence.<n>This structure is locally optimal near the point of convergence, so RAMDA is guaranteed to obtain the best structure possible.
  arXiv  Detail & Related papers  (2024-03-21T13:43:49Z)
• Transforming Image Super-Resolution: A ConvFormer-based Efficient Approach [58.57026686186709]
  We introduce the Convolutional Transformer layer (ConvFormer) and propose a ConvFormer-based Super-Resolution network (CFSR) CFSR inherits the advantages of both convolution-based and transformer-based approaches. Experiments demonstrate that CFSR strikes an optimal balance between computational cost and performance.
  arXiv  Detail & Related papers  (2024-01-11T03:08:00Z)
• Energy-efficient Task Adaptation for NLP Edge Inference Leveraging Heterogeneous Memory Architectures [68.91874045918112]
  adapter-ALBERT is an efficient model optimization for maximal data reuse across different tasks. We demonstrate the advantage of mapping the model to a heterogeneous on-chip memory architecture by performing simulations on a validated NLP edge accelerator.
  arXiv  Detail & Related papers  (2023-03-25T14:40:59Z)
• ASFD: Automatic and Scalable Face Detector [129.82350993748258]
  We propose a novel Automatic and Scalable Face Detector (ASFD) ASFD is based on a combination of neural architecture search techniques as well as a new loss design. Our ASFD-D6 outperforms the prior strong competitors, and our lightweight ASFD-D0 runs at more than 120 FPS with Mobilenet for VGA-resolution images.
  arXiv  Detail & Related papers  (2020-03-25T06:00:47Z)
• Algorithm-hardware Co-design for Deformable Convolution [40.50544352625659]
  We build an efficient object detection network with modified deformable convolutions and quantize the network using state-of-the-art quantization methods. Preliminary experiments show that little accuracy is compromised and speedup can be achieved with our co-design optimization for the deformable convolution.
  arXiv  Detail & Related papers  (2020-02-19T01:08:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.