bench --roto per size: no larger size tracked better on the synthetic clips and
Large lost the object after it re-entered; the hints say what was measured.
The bar shows the current object's stroke count and can remove its strokes.
An env-gated test runs the in-app download of a size.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Strokes carry an object (0 for older projects, nothing written for 0). The
tracker follows every object in one pass and the mattes are united; the
confidence of a frame is the least sure object's. Object picker in the Layer
panel's bar; roto.options/stroke/clearStrokes/status take or report objects.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Halves the weight memory (Large: about 900 MB to 450 MB). Off by default: the
video logit error grows from about 0.003 to 0.06-0.12 (masks unchanged) for a
speed gain of under 10 percent.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Hiera configs and size detection from the weights, registry entries pinned to
commits with verified SHA-256, the loader picks the size by checksum. The GPU
encoder submits once per block so Large fits in memory, and uploads no longer
panic on a lost device. Reference tests run per size on the CPU and GPU.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
egui_kittest test: a scripted tracker (sure, doubtful, unsure, absent, sure again)
tracks a layer and the Layer panel is rendered; the strip's pixels are green, amber,
red, slate and green at the right frames. ROTO_SPAN_SNAPSHOT saves the panel image.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
--clip FRAMES_DIR --truth MASKS_DIR runs the trackers on image sequences with true
mattes (strokes made from the truth; nothing is bundled or copied), tested end to end.
Trackers report the most memory their banks held (peakMB column).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Simulated user paints a foreground stroke through the true object and a background
stroke on the biggest false positive on the first frame under IoU 0.9, up to 6
rounds. SAM 2 needs none on every clip; classical and MobileSAM need more than 6 on
occlusion and re-entry. All trackers now take strokes on any frame.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Memory attention (1.8e-5 vs PyTorch) and the memory encoder (2.6e-6) as wgpu kernels,
convolution and layer-norm+GELU kernels, VideoModel::load_auto picks the GPU when
there is one (EFFECTCRAFT_SAM2_DEVICE overrides) and falls back to the CPU on any GPU
error; disable_gpu switches it off mid-clip. One process-wide device (several at once
crashed the Vulkan loader). A tracked frame: 4.3 s CPU, 0.7 s on the RX 7900 GRE.
Clips tracked on the GPU match Sam2VideoModel (worst logit error 0.0032).
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Strided batched matmul with bias, GELU/ReLU and residual in the epilogue, layer
norm, softmax, window partition, pooling, patch embedding, RoPE and neck kernels
(WGSL, all tested against CPU definitions). GpuEncoder keeps activations on the
GPU and returns the 64x64 level and the decoder's skip features. On the RX 7900
GRE one frame encodes in 0.29 s (debug build) against 2.1 s on the CPU in release;
errors 6.8e-6 vs PyTorch, 1.2e-5 vs the CPU path.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Candle adds a bias on one thread; the Linear layers now add it (and apply GELU
for the MLP in) in one parallel pass. Encode 2.4 s -> 2.1 s, memory attention
2.4 -> 2.3 s; reference tests unchanged. gpu_matmul example: a tiled WGSL
matmul on the RX 7900 GRE reaches 1.9-3.6 TFLOPS against 0.16-0.47 for the CPU.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A tracked frame is the encode (2.4 s) plus memory attention (2.4 s with a full
bank), about 5 s on 16 threads, not one encode as the docs said. Docs corrected;
chunked attention tried and reverted.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Sine position encodings depend only on the sizes: built once (150 ms of a frame).
do_pool takes the maximum over grouped cells instead of permute + max_pool2d +
permute (tested equal, odd sizes included). Encode 2.6 s -> 2.4 s per frame in
release on 16 threads (3.4 s at the start of the speed work); reference tests
unchanged. Chunked global attention was tried and did not help.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Candle's gelu_erf runs on one thread and was a third of the Hiera block time.
A rayon CustomOp1 calls the same erf per element (bit-identical, tested). All
reference tests unchanged. op_bench example times the block's operations.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
MaskModel::idle_secs (SAM 2 tracks its use); the engine frees a model idle for
ten minutes when no job runs, keeps the choice, and reloads it before a stroke
or a Propagate needs it. Docs: memory and limits.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Kristian Myram