Search papers, labs, and topics across Lattice.
BoltNet is an ultra-lightweight convolutional network designed for on-device plant species identification, addressing the challenges of large taxonomic label spaces and resource constraints in field applications. By integrating a Spatial Redistribution Bottleneck and Logit PreSampling, BoltNet achieves a remarkable F1-score of 0.682 with only 341K parameters, outperforming other models under 2 MB while maintaining efficiency across various hardware platforms. Comprehensive evaluations on devices like the Raspberry Pi 5 and Jetson Orin Nano demonstrate BoltNet's superior performance in terms of frames per second per watt (FPS/W), making it a compelling choice for real-time applications in plant identification.
BoltNet achieves state-of-the-art plant species identification accuracy with a fraction of the model size, making it ideal for on-device deployment.
Automated plant species identification from citizen-science imagery is an established, demanding fine-grained recognition problem: large taxonomic label spaces, visually similar species, and long-tailed observations require real model capacity, while field use constrains memory, latency, and power. Model size is only part of the deployment cost: intermediate activations held in memory during inference and platformdependent execution behavior matter too, so compact recognition must be assessed on target hardware rather than through complexity metrics alone. We present BoltNet, an ultra-lightweight fully convolutional architecture combining a Spatial Redistribution Bottleneck and Logit PreSampling to improve the tradeoff between predictive performance and model size in high-cardinality classification, and report the AccuracyCompression Tradeoff as a complementary diagnostic. On Pl@ntNet300K, BoltNet reaches 0.682 F1-score with 341K parameters (1.37 MB), the highest F1-score among evaluated models below 2 MB and close to substantially larger convolutional backbones. Model-only measurements on a Raspberry Pi 5, Jetson Orin Nano, and Hailo-8 characterize execution across CPU, GPU, and NPU platforms, where BoltNet is the most consistently efficient model, with the best FPS/W on the GPU and NPU and second-best on the CPU. Results on AIDERv2 and CLRS provide secondary evidence of transfer across environmental image-classification tasks. Code available at: https://codeberg.org/danielrossi/BoltNet
