Cross-Architecture Auxiliary Feature Space Translation for Efficient Few-Shot Personalized Object Detection

• URL: http://arxiv.org/abs/2407.01193v1
• Date: Mon, 1 Jul 2024 11:33:53 GMT
• Title: Cross-Architecture Auxiliary Feature Space Translation for Efficient Few-Shot Personalized Object Detection
• Authors: Francesco Barbato, Umberto Michieli, Jijoong Moon, Pietro Zanuttigh, Mete Ozay,
• Abstract summary: We propose an instance-level personalized object detection strategy called AuXFT. We show that AuXFT reaches a performance of 80% its upper bound at just 32% of the inference time. We validate AuXFT on three publicly available datasets and one in-house benchmark designed for the IPOD task.
• Score: 28.06000586370357
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent years have seen object detection robotic systems deployed in several personal devices (e.g., home robots and appliances). This has highlighted a challenge in their design, i.e., they cannot efficiently update their knowledge to distinguish between general classes and user-specific instances (e.g., a dog vs. user's dog). We refer to this challenging task as Instance-level Personalized Object Detection (IPOD). The personalization task requires many samples for model tuning and optimization in a centralized server, raising privacy concerns. An alternative is provided by approaches based on recent large-scale Foundation Models, but their compute costs preclude on-device applications. In our work we tackle both problems at the same time, designing a Few-Shot IPOD strategy called AuXFT. We introduce a conditional coarse-to-fine few-shot learner to refine the coarse predictions made by an efficient object detector, showing that using an off-the-shelf model leads to poor personalization due to neural collapse. Therefore, we introduce a Translator block that generates an auxiliary feature space where features generated by a self-supervised model (e.g., DINOv2) are distilled without impacting the performance of the detector. We validate AuXFT on three publicly available datasets and one in-house benchmark designed for the IPOD task, achieving remarkable gains in all considered scenarios with excellent time-complexity trade-off: AuXFT reaches a performance of 80% its upper bound at just 32% of the inference time, 13% of VRAM and 19% of the model size.

Related papers

• OccLoff: Learning Optimized Feature Fusion for 3D Occupancy Prediction [5.285847977231642]
  3D semantic occupancy prediction is crucial for ensuring the safety in autonomous driving. Existing fusion-based occupancy methods typically involve performing a 2D-to-3D view transformation on image features. We propose OccLoff, a framework that Learns to optimize Feature Fusion for 3D occupancy prediction.
  arXiv  Detail & Related papers  (2024-11-06T06:34:27Z)
• Scale-Invariant Feature Disentanglement via Adversarial Learning for UAV-based Object Detection [18.11107031800982]
  We propose to improve single-stage inference accuracy through learning scale-invariant features. We apply our approach to three state-of-the-art lightweight detection frameworks on three benchmark datasets.
  arXiv  Detail & Related papers  (2024-05-24T11:40:22Z)
• ESOD:Edge-based Task Scheduling for Object Detection [8.347247774167109]
  We present a novel edge-based task scheduling framework for object detection (termed as ESOD) The results show that ESOD can reduce latency and energy consumption by an average of 22.13% and 29.60%.
  arXiv  Detail & Related papers  (2021-10-20T13:43:51Z)
• Efficient Person Search: An Anchor-Free Approach [86.45858994806471]
  Person search aims to simultaneously localize and identify a query person from realistic, uncropped images. To achieve this goal, state-of-the-art models typically add a re-id branch upon two-stage detectors like Faster R-CNN. In this work, we present an anchor-free approach to efficiently tackling this challenging task, by introducing the following dedicated designs.
  arXiv  Detail & Related papers  (2021-09-01T07:01:33Z)
• When Liebig's Barrel Meets Facial Landmark Detection: A Practical Model [87.25037167380522]
  We propose a model that is accurate, robust, efficient, generalizable, and end-to-end trainable. In order to achieve a better accuracy, we propose two lightweight modules. DQInit dynamically initializes the queries of decoder from the inputs, enabling the model to achieve as good accuracy as the ones with multiple decoder layers. QAMem is designed to enhance the discriminative ability of queries on low-resolution feature maps by assigning separate memory values to each query rather than a shared one.
  arXiv  Detail & Related papers  (2021-05-27T13:51:42Z)
• Analysis of voxel-based 3D object detection methods efficiency for real-time embedded systems [93.73198973454944]
  Two popular voxel-based 3D object detection methods are studied in this paper. Our experiments show that these methods mostly fail to detect distant small objects due to the sparsity of the input point clouds at large distances. Our findings suggest that a considerable part of the computations of existing methods is focused on locations of the scene that do not contribute with successful detection.
  arXiv  Detail & Related papers  (2021-05-21T12:40:59Z)
• Learnable Online Graph Representations for 3D Multi-Object Tracking [156.58876381318402]
  We propose a unified and learning based approach to the 3D MOT problem. We employ a Neural Message Passing network for data association that is fully trainable. We show the merit of the proposed approach on the publicly available nuScenes dataset by achieving state-of-the-art performance of 65.6% AMOTA and 58% fewer ID-switches.
  arXiv  Detail & Related papers  (2021-04-23T17:59:28Z)
• Unsupervised and self-adaptative techniques for cross-domain person re-identification [82.54691433502335]
  Person Re-Identification (ReID) across non-overlapping cameras is a challenging task. Unsupervised Domain Adaptation (UDA) is a promising alternative, as it performs feature-learning adaptation from a model trained on a source to a target domain without identity-label annotation. In this paper, we propose a novel UDA-based ReID method that takes advantage of triplets of samples created by a new offline strategy.
  arXiv  Detail & Related papers  (2021-03-21T23:58:39Z)
• It's always personal: Using Early Exits for Efficient On-Device CNN Personalisation [19.046126301352274]
  On-device machine learning is becoming a reality thanks to the availability of powerful hardware and model compression techniques. In this work, we observe that a much smaller, personalised model can be employed to fit a specific scenario. We introduce PershonEPEE, a framework that attaches early exits on the model and personalises them on-device.
  arXiv  Detail & Related papers  (2021-02-02T09:10:17Z)
• SA-Det3D: Self-Attention Based Context-Aware 3D Object Detection [9.924083358178239]
  We propose two variants of self-attention for contextual modeling in 3D object detection. We first incorporate the pairwise self-attention mechanism into the current state-of-the-art BEV, voxel and point-based detectors. Next, we propose a self-attention variant that samples a subset of the most representative features by learning deformations over randomly sampled locations.
  arXiv  Detail & Related papers  (2021-01-07T18:30:32Z)
• SensiX: A Platform for Collaborative Machine Learning on the Edge [69.1412199244903]
  We present SensiX, a personal edge platform that stays between sensor data and sensing models. We demonstrate its efficacy in developing motion and audio-based multi-device sensing systems. Our evaluation shows that SensiX offers a 7-13% increase in overall accuracy and up to 30% increase across different environment dynamics at the expense of 3mW power overhead.
  arXiv  Detail & Related papers  (2020-12-04T23:06:56Z)

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.