Ayase Models

--- tags: - ayase - evaluation - runtime-assets --- # Ayase Runtime Assets Versioned runtime assets consumed by [Ayase](https://github.com/seruva19/ayase) evaluation modules. ## Contents - pretrained checkpoints from upstream metric implementations; - pinned source archives for evaluators without stable package releases; - inference configurations, calibration data, and support files. Assets are grouped by Ayase module or evaluator, for example: ```text advanced_flow/ brightvq/ cgvqm/ dover/ fast_vqa/ expression_following/ mj_video/ uvq1p5/ vbench2/ vebench/ vqa2/ worldmodelbench/ ``` Ayase downloads only the paths required by selected modules. Release-critical consumers pin a repository revision; selected modules also validate file size or SHA-256 before loading. Current logical size: approximately **25.33 GiB across 107 files**. ## Provenance and licenses Files originate from multiple upstream projects and retain their respective licenses and usage terms. Source and license metadata are stored alongside assets where available; consult the corresponding upstream project before redistribution.

--- license: "mit" --- # Ayase Models Pre-downloaded model weights for [Ayase](https://github.com/seruva19/ayase) modules that require non-HuggingFace downloads. These models are hosted here because their original CDNs (dl.fbaipublicfiles.com, openaipublic.azureedge.net, GitHub releases) can be unreliable or blocked on some servers. ## Structure ``` ayase-models/ ├── dover/ # DOVER video quality assessment (ICCV 2023) │ ├── DOVER.pth # 229 MB — S-Lab License 1.0 │ └── convnext_tiny_1k_224_ema.pth # 110 MB — MIT ├── i2v_similarity/ # Image-to-Video similarity metrics │ ├── ViT-B-32.safetensors # 338 MB — MIT (OpenAI CLIP) │ ├── dinov2_vitb14_pretrain.pth # 331 MB — Apache 2.0 (Meta) │ └── alex.pth # 6 KB — BSD-2 (LPIPS) ├── dino_face_identity/ # DINOv2 face-crop identity indicator │ └── dinov2_vitb14_pretrain.pth # 331 MB — Apache 2.0 (Meta) ├── dreamsim/ # DreamSim perceptual similarity — base DINO backbone │ └── dino_vitbase16_pretrain.pth # 327 MB — Apache 2.0 (Meta, DINO) ├── advanced_flow/ # RAFT optical flow (ECCV 2020) │ ├── raft_large_C_T_SKHT_V2-ff5fadd5.pth # 21 MB — BSD-3 │ └── raft_small_C_T_V2-01064c6d.pth # 3.9 MB — BSD-3 ├── fast_vqa/ # FAST-VQA quality assessment (ECCV 2022) │ ├── FAST_VQA_3D_1_1.pth # 121 MB — MIT │ ├── FAST_VQA_B_1_4.pth # 121 MB — MIT │ └── FAST_VQA_M_1_4.pth # 105 MB — MIT ├── aesthetic_scoring/ # LAION aesthetic predictor │ └── sac+logos+ava1-l14-linearMSE.pth # 3.5 MB — MIT ├── video_memorability/ # Video memorability estimation │ └── dinov2_vits14_pretrain.pth # 84 MB — Apache 2.0 (Meta) ├── spectral/ # Spectral analysis │ └── dinov2_vits14_pretrain.pth # 84 MB — Apache 2.0 (Meta) ├── trajan/ # Point tracking (CoTracker2) │ └── cotracker2.pth # 194 MB — Apache 2.0 (Meta) ├── depth_map_quality/ # Monocular depth estimation │ ├── dpt_swin2_tiny_256.pt # 164 MB — MIT (Intel ISL MiDaS) │ └── midas_v21_small_256.pt # 82 MB — MIT (Intel ISL MiDaS) ├── depth_consistency/ # Temporal depth consistency │ ├── dpt_swin2_tiny_256.pt # 164 MB — MIT │ └── midas_v21_small_256.pt # 82 MB — MIT ├── motion_smoothness/ # RIFE motion smoothness │ └── flownet.pkl # Motion interpolation network ├── brightvq/ # BrightRate / BrightVQ HDR no-reference quality │ ├── brightrate_brightvq.pt # 161 MB — BrightRate regressor │ ├── CONTRIQUE_checkpoint25.tar # 107 MB — CONTRIQUE feature extractor │ ├── frames_modelparameters.mat # 8 KB — NIQE/H...

Source context: 0 downloads · 0 likes · Repo AkaneTendo25/ayase-runtime-assets