AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

AnimalClue: Recognizing Animals by their Traces

Risa Shinoda, Nakamasa Inoue, Iro Laina, Christian Rupprecht, Hirokatsu Kataoka

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Remote sensing segmentation and change · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add taxonomic context (genus, family, order, class) to CLIP text prompts (ablation)fine-tuned CLIP · AnimalClue footprint species classificationtop-1 accuracy: 17.6% to 21.7%Improved
use Swin-B instead of ResNet-50classifier · AnimalClue feather species classificationtop-1 accuracy (%): 59.7 -> 65.3Improved
use Swin-B instead of VGG-16classifier · AnimalClue bone species classificationtop-1 accuracy (%): 14.7 -> 20.5Improved
restrict evaluation to frequent species categoriesSwin-B · AnimalClue egg species classificationtop-1 accuracy (%): 49.4 -> 58.5Improved
evaluate on rare species categories instead of allSwin-B · AnimalClue feather species classificationtop-1 accuracy (%): 65.3 -> 2.52Hurt
evaluate on rare species categories instead of allRT-DETR · AnimalClue feather species detectionmAP@50-95: 0.17 -> 0.00Hurt
use RT-DETR instead of YOLOv11detector · AnimalClue egg species detectionmAP@50-95: 0.14 -> 0.04Hurt
use RT-DETR instead of Faster-RCNNdetector · AnimalClue feces species detectionmAP@50-95: 0.06 -> 0.17Improved
use MaskDINO instead of Mask-RCNNsegmenter · AnimalClue egg species instance segmentationmask mAP@50-95: 0.16 -> 0.25Improved
fine-tune CLIP on AnimalClue instead of using generic CLIP or BioCLIPCLIP ViT-B-32 · AnimalClue feature separation (t-SNE)qualitative feature separationImproved
use BioCLIP instead of standard CLIPCLIP features · AnimalClue feather vs egg separation (t-SNE)qualitative feature separationImproved
predict coarser taxonomic level (order instead of species)Swin-B · AnimalClue footprint classificationtop-1 accuracy (%): 32.3 -> 66.1Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy