🔄 Self-Supervised Learning

🎞️ ECCV2026 · 5 paper notes

📌 Same area in other venues: 📷 CVPR2026 (92) · 🔬 ICLR2026 (81) · 💬 ACL2026 (1) · 🧪 ICML2026 (28) · 🤖 AAAI2026 (16) · 🧠 NeurIPS2025 (35)

Bottom-up modeling of repeated elements via single image analysis-by-synthesis

A shared two-dimensional prototype with adjustable appearance, single-image crop-based curriculum optimization, and full-image selection and refinement turn repeated-element discovery into interpretable reconstruction; on 116 curated FSC-147 images, the method achieves an image PSNR of 23.76 and a counting MAE of 14.11, without implying that the entire system is free of pretraining or scale priors.

DP-BOA: Dirichlet-Process Birth-or-Assign for On-the-Fly Category Discovery

DP-BOA reformulates per-sample online category discovery as a Bayesian comparison between existing-category assignment evidence and new-category birth evidence; support-set prior calibration, full-covariance prediction, and online statistical updates yield 53.4% overall and 51.6% novel-class clustering accuracy on CUB, without leading on every dataset or known-class metric.

ExPLoRe: Expert Patch-Level Loss Routing for Multi-Objective Masked Image Modeling

ExPLoRe repurposes the dispatch weights of Soft-MoE as learnable, patch-level loss coefficients in multi-objective masked image modeling. Through a loss-coupling mechanism, the router dynamically adjusts the weights of different objectives based on the semantic properties of different image regions. On ImageNet-1K, it achieves 80.6% linear probing accuracy and 85.3% fine-tuning accuracy with lower inference computational cost (11.93 GFLOPs vs. 17.45 GFLOPs). It also introduces three downstream transfer strategies (Freeze Routing / Expert Dropout / Freeze Attention) that completely close the 2.5–2.9 mIoU gap in semantic segmentation.

From Phase to Phenomenon: Self-Supervised Learning of Subsurface Scattering with Minimal Phase-shift Inputs

This work proposes a self-supervised pre-training framework that learns general subsurface scattering (SSS) representations using a projector-camera system from only 8 high-frequency phase-shift profilometry (PSP) images. It then predicts anisotropic scattering footprints via a decoder to achieve zero-shot relighting, reducing the data acquisition requirements for new objects from thousands of images to just 8.

LoT-Pass: Long-term-robust Image Watermarking for Image to Video Generation

LoT-Pass proposes a bidirectional robustness strategy: simulating temporal evolution during training to actively expand the watermarking robustness distance, and employing optical flow inversion during inference to "pull back" severely drifted frames into an extractable state. This achieves a video-level watermark extraction accuracy of over 95% for the first time in image-to-video (I2V) generation scenarios.