Yicheng Zhang, Yujie Jin, Ke Zhang, Jiaxin Wu · 2026 · arXiv
DOI: 10.48550/arXiv.2610.28590 · open-access PDF
Gambar-gambar kamera-trap menyajikan variasi substansial dalam penampilan hewan, pencahayaan, kekacauan latar belakang, dan degradasi visual. Autoencoders yang disamarkan (MAEs) menawarkan kerangka kerja bebas label untuk memodelkan data tersebut melalui rekonstruksi patch gambar yang tersembunyi secara acak. Kami menyajikan WildMAE, ekstensi dekoder-side dari MAE yang menggabungkan perbaikan struktur lokal yang sadar dengan perhatian silang yang disamarkan ke yang terlihat. Jalur residual konvolusi nol gerbang memperkenalkan lingkungan dua dimensi sebelumnya, sementara perhatian silang gerbang memungkinkan posisi disamarkan untuk mendapatkan informasi kontekstual dari pos yang terlihat.
Machine translation (Meta M2M100); the English original follows.
Camera-trap imagery presents substantial variation in animal appearance, illumination, background clutter, and visual degradation. Masked autoencoders (MAEs) offer a label-free framework for modeling such data through reconstruction of randomly hidden image patches. We present WildMAE, a decoder-side extension of MAE that combines local structure-aware refinement with masked-to-visible cross-attention. A zero-gated convolutional residual path introduces a two-dimensional neighborhood prior, while gated cross-attention allows masked positions to retrieve contextual information from visible-posi
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28590, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange