Views
No views yet
.tflite via the official litert_torch path. ×4 real-world super-resolution.
FP16, 3.5 MB, input 128×128 → output 512×512 (NHWC, RGB, 0–1).GREATER+SELECT+MUL
and PixelShuffle lowers to a >4-D reshape — both GPU-rejected. Here:relu(x) − a·relu(−x) (per-channel a): exact, only RELU/MUL/SUB.TRANSPOSE_CONV, no >4-D tensors.[1, 128, 128, 3] NHWC, RGB, 0–1 float (no mean/std). Tile larger images into
128×128 patches.[1, 512, 512, 3] NHWC, RGB, 0–1 (clamp to [0,1]). ×4 upscale.realesr-general-x4v3 was trained by the Real-ESRGAN authors on public super-resolution datasets
(DIV2K / Flickr2K / OST and a synthetic high-order degradation pipeline). The model upscales image
pixels only — no faces, identities, or other personal attributes are detected, recognized, or
output. No additional or private data was used; weights are the official release, only the op graph
was re-authored for GPU.litert_torch conversion script:
https://github.com/google-ai-edge/litert-samples (compiled_model_api/super_resolution)