A typical 2D-to-3D pipeline takes multi-view images as input, where a Vision Foundation Model (VFM) extracts features that are spatially upsampled to dense representations for 3D reconstruction. If dense features across views preserve geometric consistency, differentiable rendering can recover an accurate 3D representation, making the feature upsampler a critical component. Recent learnable upsampling methods mainly aim to enhance spatial details, such as sharper geometry or richer textures, yet their impact on 3D awareness remains underexplored. To address this gap, we introduce a spectral diagnostic framework with six complementary metrics that characterize amplitude redistribution, structural spectral alignment, and directional stability. Across classical interpolation and learnable upsampling methods on CLIP and DINO backbones, we observe three key findings. First, structural spectral consistency (SSC/CSC) is the strongest predictor of NVS quality, whereas High-Frequency Spectral Slope Drift (HFSS) often correlates negatively with reconstruction performance, indicating that emphasizing high-frequency details alone does not necessarily improve 3D reconstruction. Second, geometry and texture respond to different spectral properties: Angular Energy Consistency (ADC) correlates more strongly with geometry-related metrics, while SSC/CSC influence texture fidelity slightly more than geometric accuracy. Third, although learnable upsamplers often produce sharper spatial features, they rarely outperform classical interpolation in reconstruction quality, and their effectiveness depends on the reconstruction model. Overall, our results indicate that reconstruction quality is more closely related to preserving spectral structure than to enhancing spatial detail, highlighting spectral consistency as an important principle for designing upsampling strategies in 2D-to-3D pipelines.
MAPS: A Synthetic Dataset for Probing Vision Models in a Controlled 3D Scene Space
FIAS & Institute of Computer Science Goethe University Frankfurt(FIAS与计算机科学研究所弗赖堡大学);Mila & Department of Biology York University(Mila与生物学系约克大学);Institute of Computer Science Goethe University Frankfurt(计算机科学研究所弗赖堡大学)
SceneBench: A Hierarchical Benchmark for Vision-Language Understanding of 3D Scenes
Nepal Applied Mathematics and Informatics Institute for research (NAAMII)(尼泊尔应用数学与信息学研究所);INSAIT, Sofia University(索非亚大学INSAIT);State University of New York at Stony Brook(纽约州立大学石溪分校)
Probing neural representations of scene perception in a hippocampally dependent task using artificial neural networks
Kavli Institute for Systems Neuroscience(卡夫利系统神经科学研究所);NTNU(挪威科技大学);Max-Planck-Insitute for Human Cognitive and Brain Sciences(马克斯·普朗克人类认知与脑科学研究所);UCL(伦敦大学学院)
HumanPCR: Probing MLLM Capabilities in Diverse Human-Centric Scenes
Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of Computing Technology, CAS, China(中国科学院智能信息处理重点实验室(中国科学院)、计算技术研究所、中国)
Automatic Generation of Contrast Sets from Scene Graphs: Probing the Compositional Consistency of GQA
School of Computer Science and Engineering, The Hebrew University of Jerusalem(耶路撒冷希伯来大学计算机科学与工程学院);Department of Computer Science, Ben Gurion University(本·古里安大学计算机科学系)
Beyond Pixels: A Differentiable Pipeline for Probing Neuronal Selectivity in 3D
Institute for Computer Science, University of Göttingen(盖尔廷根大学计算机科学研究所);Campus Institute Data Science (CIDAS)(校园数据科学研究所);Lower Saxony Center for AI & Causal Methods in Medicine(下萨克森人工智能与因果医学中心);International Max Planck Research School for Intelligent Systems(国际马克斯·普朗克智能系统研究学校);Noselab GmbH(诺赛拉德公司)
SpatialAct: Probing Spatial Reasoning-to-Action Capabilities of VLM Agents in 3D Scenes
The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州));Zhongguancun Academy(中关村学院);Tsinghua University(清华大学);Helsinki University(赫尔辛基大学)
Huazhong University of Science and Technology(华中科技大学);School of Artificial Intelligence and Automation, Huazhong University of Science and Technology(华中科技大学人工智能与自动化学院)
Is VLA Reasoning Faithful? Probing Safety of Chain-of-Causation in Autonomous Driving Models
School of Computer Science and Engineering(计算机科学与工程学院);Central South University(中南大学);School of Computer Science(计算机科学学院);University of Wollongong in Dubai(迪拜大学)
CycleSAM: Few-Shot Surgical Scene Segmentation with Cycle- and Scene-Consistent Feature Matching
University of Strasbourg, CNRS, INSERM, ICube, UMR7357(斯特拉斯堡大学、国家科学研究中心、法国国家医学研究院、ICube、UMR7357);IHU Strasbourg(斯特拉斯堡IHU);Fondazione Policlinico Universitario A. Gemelli IRCCS(阿·吉梅利大学附属医院基金会)