Lightweight Distillation of SAM 3 and DINOv3 for Edge-Deployable Individual-Level Livestock Monitoring and Longitudinal Visual Analytics

• URL: http://arxiv.org/abs/2604.27128v1
• Date: Wed, 29 Apr 2026 19:25:22 GMT
• Title: Lightweight Distillation of SAM 3 and DINOv3 for Edge-Deployable Individual-Level Livestock Monitoring and Longitudinal Visual Analytics
• Abstract summary: Foundation-model pipelines for individual-level livestock monitoring have raised the accuracy ceiling of precision livestock farming.<n>But their GPU memory budgets exceed the envelope of commodity edge accelerators.<n>To close this gap, the backbone of SAM 3 is distilled into a 40.66M-scale multi-scale student.<n>The pipeline achieves 97.34% top-1 accuracy with 91.67% macro-F1 on nine-class pig behaviour classification.
• Score: 0.3609538870261841
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Foundation-model pipelines for individual-level livestock monitoring -- combining open-vocabulary detection, promptable video segmentation, and self-supervised visual embeddings -- have raised the accuracy ceiling of precision livestock farming (PLF), but their GPU memory budgets exceed the envelope of commodity edge accelerators. To close this gap, the 446M-parameter Perception Encoder (PE-ViT-L+) backbone of SAM 3 is distilled into a 40.66M-parameter multi-scale student through three mechanisms: a Feature Pyramid Network student encoder built on TinyViT-21M-512, a four-term direction-then-scale distillation loss, and backbone-substitution inference with sliding-window session pruning that bounds streaming GPU memory growth. The DINOv3 family includes a pre-distilled ViT-S/16 variant (21.6M parameters) released alongside a 6716M-parameter ViT-7B teacher; the ViT-S (21M) variant is adopted as the per-individual embedder. On the Edinburgh Pig dataset, the compressed pipeline reaches 92.29% MOTA and 96.15% IDF1 against the SAM 3 teacher (1.68- and 0.84-percentage-point losses), achieves a 7.77-fold reduction in system-level parameters and a 3.01-fold reduction in peak VRAM (19.52GB -> 6.49GB), and reaches 97.34% top-1 accuracy with 91.67% macro-F1 on nine-class pig behaviour classification. The pipeline fits inside an NVIDIA Jetson Orin NX 16GB envelope with 4.9GB of headroom, supporting a proposed -- but not yet empirically validated -- on-device embedding-pool re-identification mechanism whose per-individual footprint of approximately 94MB per animal per year produces a longitudinal visual record amenable to retrospective association with disease, lameness, reproductive, and growth outcome labels.

Related papers

• WaVeFuse: Regime-Adaptive Equity Index Forecasting via Channel-Wise Wavelet Denoising and Vertical Attention Fusion [0.0]
  WaVeFuse is a hybrid Deep Learning for equity forecasting framework.<n>It addresses propagation of technical indicators (TIs), channel-indiscriminate decomposition, and static multi-branch fusion.<n>WaVeFuse delivers a computationally efficient, regime-robust framework suitable for research and decision-support deployment.
  arXiv  Detail & Related papers  (2026-09-13T18:50:33Z)
• On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability [57.03393882933515]
  We describe the architecture and ablations of Qwen3.8-Flash-Next, a sparse mixture-of-experts model with 125B parameters, 6B activated per token, and 51B parameters of n-gram embedding tables held off the accelerator.<n>On fourteen pre-training benchmarks the model leads the 397B-A17B predecessor on eight and trails it on the rest by at most 2.6 points, at 1/3 the activated parameters, 1/3 the training tokens, and roughly 1/9 the training FLOPs.
  arXiv  Detail & Related papers  (2026-08-31T06:35:07Z)
• Lightweight Image Classification of Raptor Species for Edge Devices: Rare-Species Dataset Expansion via Video Frame Extraction, Knowledge Distillation, and TensorRT Deployment [0.060917028769172786]
  We investigate lightweight raptor-species classification for real-time edge deployment in wind-turbine collision mitigation.<n>Using DINOv2-L (304M parameters) as a teacher, we distilled three lightweight students (MobileNetV4, ViT-Small, and EfficientNet-B0).<n>To reduce confusion between closely related species, we expanded the dataset to 2,050 images.<n>In five-seed controlled comparisons, neither distillation (versus CE-only) nor the change from a DINOv2-L to a DINOv3-L teacher yielded a clear ensemble-level improvement.
  arXiv  Detail & Related papers  (2026-07-28T20:19:25Z)
• CLEAR-MoE: Shared-Basis Expert Extraction from Frozen Vision Transformers via Calibration-Driven Layer Selection [0.0]
  CLEAR-MoE is a post-training pipeline that converts a frozen pretrained Vision Transformer (ViT) into a sparse Mixture-of-Experts (MoE) model without updating backbone weights.<n>On a GTX 960, routing and scatter-gather overhead make the CLEAR-MoE FFN 1.3-1.7x slower than the dense routers.
  arXiv  Detail & Related papers  (2026-06-26T18:12:03Z)
• LocalNav: Distilling Frontier VLMs and Embodied RL for On-Device Object Goal Navigation [53.88286427894783]
  Vision Language Models (VLMs) have emerged in the robotic domain as a powerful tool that enables environmental perception with language context.<n>Yet, their computational footprint typically confines them to cloud execution, hindering low-latency inference with local deployment on resource-constrained robots.<n>We present a distillation strategy that transfers complex spatial-semantic reasoning from large frontier models into a lightweight, 4B- parameter local VLM for edge execution on embedded GPU devices.
  arXiv  Detail & Related papers  (2026-06-26T09:11:59Z)
• EdgeSpike: Spiking Neural Networks for Low-Power Autonomous Sensing in Edge IoT Architectures [0.0]
  EdgeSpike is a co-designed spiking neural network (SNN) framework for autonomous low-power sensing in edge Internet of Things (IoT) architectures.<n>It unifies a hybrid surrogate-gradient and direct-encoding training pipeline, a hardware-aware neural architecture search, and an event-driven runtime.<n>It achieves a mean classification accuracy of 91.4%, within 1.2 percentage points (pp) of strong INT8 convolutional neural network (CNN) baselines.
  arXiv  Detail & Related papers  (2026-04-29T05:15:28Z)
• ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems [51.56484100374058]
  We present ZenBrain, a multilayer memory architecture integrating neuroscience models.<n>It implements seven memory layers (working, short-term, episodic, semantic, procedural, core, foundational cross-context) orchestrated by nine algorithms.<n>On LongMemEval, ZenBrain holds the highest rank on all 12 system judges.
  arXiv  Detail & Related papers  (2026-04-26T20:39:19Z)
• SpaCeFormer: Fast Proposal-Free Open-Vocabulary 3D Instance Segmentation [82.87586897359367]
  SpaCeFormer is a proposal-free space-curve transformer that runs at 0.14 seconds per scene.<n>SpaCeFormer reaches 21x higher mask recall than prior single-view pipelines.<n>On ScanNet200 we achieve 11.1 zero-shot mAP, a 2.8x improvement over the prior best proposal-free method.
  arXiv  Detail & Related papers  (2026-04-22T09:57:57Z)
• Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory [76.63021613850093]
  We deploy an autonomous research pipeline to discover Omni-SimpleMem, a unified multimodal memory framework for lifelong AI agents.<n>System achieves state-of-the-art on two benchmarks, improving F1 by +411% on LoCoMo and +214% on Mem-Gallery.<n>We provide a taxonomy of six discovery types and identify four properties that make multimodal memory particularly suited for autoresearch.
  arXiv  Detail & Related papers  (2026-04-01T15:06:23Z)
• A Novel Compression Framework for YOLOv8: Achieving Real-Time Aerial Object Detection on Edge Devices via Structured Pruning and Channel-Wise Distillation [0.0]
  We propose a novel three-stage compression pipeline for the YOLOv8 object detection model.<n>Sparsity-aware training, structured channel pruning, and Channel-Wise Knowledge Distillation (CWD) are used.<n>Experiments on the VisDrone dataset demonstrate the effectiveness of our approach across multiple YOLOv8 variants.
  arXiv  Detail & Related papers  (2025-09-16T10:11:59Z)
• ParFormer: A Vision Transformer with Parallel Mixer and Sparse Channel Attention Patch Embedding [9.144813021145039]
  This paper introduces ParFormer, a vision transformer that incorporates a Parallel Mixer and a Sparse Channel Attention Patch Embedding (SCAPE) ParFormer improves feature extraction by combining convolutional and attention mechanisms. For edge device deployment, ParFormer-T excels with a throughput of 278.1 images/sec, which is 1.38 $times$ higher than EdgeNeXt-S. The larger variant, ParFormer-L, reaches 83.5% Top-1 accuracy, offering a balanced trade-off between accuracy and efficiency.
  arXiv  Detail & Related papers  (2024-03-22T07:32:21Z)
• It Takes Two: Masked Appearance-Motion Modeling for Self-supervised Video Transformer Pre-training [76.69480467101143]
  Self-supervised video transformer pre-training has recently benefited from the mask-and-predict pipeline. We explicitly investigate motion cues in videos as extra prediction target and propose our Masked Appearance-Motion Modeling framework. Our method learns generalized video representations and achieves 82.3% on Kinects-400, 71.3% on Something-Something V2, 91.5% on UCF101, and 62.5% on HMDB51.
  arXiv  Detail & Related papers  (2022-10-11T08:05:18Z)
• EATFormer: Improving Vision Transformer Inspired by Evolutionary Algorithm [111.17100512647619]
  This paper explains the rationality of Vision Transformer by analogy with the proven practical evolutionary algorithm (EA) We propose a novel pyramid EATFormer backbone that only contains the proposed EA-based transformer (EAT) block. Massive quantitative and quantitative experiments on image classification, downstream tasks, and explanatory experiments demonstrate the effectiveness and superiority of our approach.
  arXiv  Detail & Related papers  (2022-06-19T04:49:35Z)
• Alexa Teacher Model: Pretraining and Distilling Multi-Billion-Parameter Encoders for Natural Language Understanding Systems [63.713297451300086]
  We present results from a large-scale experiment on pretraining encoders with non-embedding parameter counts ranging from 700M to 9.3B. Their subsequent distillation into smaller models ranging from 17M-170M parameters, and their application to the Natural Language Understanding (NLU) component of a virtual assistant system.
  arXiv  Detail & Related papers  (2022-06-15T20:44:23Z)

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.