paper-with-me

홈 › Papers

Token Pruning using a Lightweight Background Aware Vision Transformer

2024-10-12 · Sudhakar Sah, Ravish Kumar, Honnesh Rohmetra, Ehsan Saboori

High runtime memory and high latency puts significant constraint on Vision Transformer training and inference, especially on edge devices. Token pruning reduces the number of input tokens to the ViT based on importance criteria of each token. We present a Background Aware Vision Transformer (BAViT) model, a pre-processing block to object detection models like DETR/YOLOS aimed to reduce runtime memory and increase throughput by using a novel approach to identify background tokens in the image. The background tokens can be pruned completely or partially before feeding to a ViT based object detector. We use the semantic information provided by segmentation map and/or bounding box annotation to train a few layers of ViT to classify tokens to either foreground or background. Using 2 layers and 10 layers of BAViT, background and foreground tokens can be separated with 75% and 88% accuracy on VOC dataset and 71% and 80% accuracy on COCO dataset respectively. We show a 2 layer BAViT-small model as pre-processor to YOLOS can increase the throughput by 30% - 40% with a mAP drop of 3% without any sparse fine-tuning and 2% with sparse fine-tuning. Our approach is specifically targeted for Edge AI use cases.

📄 PDF Abstract BibTeX arXiv:2410.09324

Code (0)

등록된 구현이 없습니다.

Tasks

object-detectionObject Detection

Methods 이 논문이 사용한 방법론

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Residual Connection 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Adam 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Vision Transformer The Vision Transformer, or ViT, is a model for image classification that employs a Transformer-like architecture over…

Similar Papers 제목 키워드 기반

Index-Preserving Lightweight Token Pruning for Efficient Document Understanding in Vision-Language Models

2025-09-08 · Jaemin Son, Sujin Choi, Inyong Yun arxiv

Recent progress in vision-language models (VLMs) has led to impressive results in document understanding tasks, but their high computational demands remain a challenge. To mitigate the compute burdens, we propose a light…

Adaptive and Background-Aware Vision Transformer for Real-Time UAV Tracking

2023-01-01 · ICCV 2023 1 · Shuiwang Li, Yangxiang Yang, Dan Zeng, Xucheng Wang

While discriminative correlation filters (DCF)-based trackers prevail in UAV tracking for their favorable efficiency, lightweight convolutional neural network (CNN)-based trackers using filter pruning have also demon…

image-classificationImage Classification

Walk and Read Less: Improving the Efficiency of Vision-and-Language Navigation via Tuning-Free Multimodal Token Pruning

2025-09-18 · Wenda Qin, Andrea Burns, Bryan A. Plummer, Margrit Betke arxiv

Large models achieve strong performance on Vision-and-Language Navigation (VLN) tasks, but are costly to run in resource-limited environments. Token pruning offers appealing tradeoffs for efficiency with minimal performa…

QAPruner: Quantization-Aware Vision Token Pruning for Multimodal Large Language Models

2026-04-03 · Xinhao Wang, Zhonyu Xia, Zhiwei Lin, Zhe Li 외 arxiv

Multimodal Large Language Models (MLLMs) have shown strong reasoning ability, but their high computational and memory costs hinder deployment in resource-constrained settings. While Post-Training Quantization (PTQ) and v…

ST-Prune: Training-Free Spatio-Temporal Token Pruning for Vision-Language Models in Autonomous Driving

2026-04-21 · Lin Sha, Haiyun Guo, Tao Wang, Cong Zhang 외 arxiv

Vision-Language Models (VLMs) have become central to autonomous driving systems, yet their deployment is severely bottlenecked by the massive computational overhead of multi-view camera and multi-frame video input. Exist…

Autonomous Driving