Struct BitrateEstimator 

pub struct BitrateEstimator { /* private fields */ }

Estimates output bitrate (bits per second) from encode parameters.

The model is a simplified empirical formula: bitrate ≈ base_bpp * pixels_per_frame * frame_rate * quality_factor

where quality_factor decreases as QP/CRF increases.

An embedded BitrateRunningAnalyzer accumulates per-frame bitrate statistics incrementally via Welford’s online algorithm so that callers can query running mean/variance without storing historical data.

Example

use oximedia_transcode::bitrate_estimator::BitrateEstimator;

let mut est = BitrateEstimator::new();
let bps = est.estimate_from_crf(23, 1920, 1080, 30.0);
assert!(bps > 0);

// Feed observed per-frame bit counts into the running analyzer
est.record_frame_bits(50_000);
est.record_frame_bits(45_000);
let summary = est.running_summary();
assert!(summary.mean_bps > 0.0);

Implementations

impl BitrateEstimator

pub fn new() -> Self

Creates a BitrateEstimator with default empirical parameters.

The embedded running analyzer is initialised for 30 fps with a 60-frame rolling window; call with_params_and_fps to customise these.

pub fn with_params(base_bpp: f64, decay: f64) -> Self

Creates a BitrateEstimator with custom parameters.

• base_bpp – bits per pixel at QP = 0.
• decay – exponential decay constant; higher = steeper quality/bitrate curve.

pub fn with_params_and_fps( base_bpp: f64, decay: f64, fps: f64, window_frames: usize, ) -> Self

Creates a BitrateEstimator with custom model parameters and analyzer config.

• fps – Frame rate for the running analyzer (bits/frame → bits/s).
• window_frames – Rolling-window size for recent-peak detection.

pub fn record_frame_bits(&mut self, bits_per_frame: u64)

Records the observed bit count for one encoded frame into the running analyzer.

Use this to track actual per-frame bitrate statistics incrementally (Welford algorithm) without buffering all historical frame data.

pub fn running_summary(&self) -> BitrateSummary

Returns a snapshot of current running bitrate statistics.

The summary reflects all frames recorded via record_frame_bits.

pub fn reset_running_stats(&mut self)

Resets the running bitrate statistics to their initial state.

pub fn estimate_from_crf( &self, crf: u8, width: u32, height: u32, frame_rate: f64, ) -> u64

Estimates output bitrate in bits/s from a CRF value (0–51 for H.264/H.265).

• crf – Constant Rate Factor (0 = lossless, 51 = worst).
• width – Frame width in pixels.
• height – Frame height in pixels.
• frame_rate – Frames per second.

pub fn estimate_from_qp( &self, qp: f64, width: u32, height: u32, frame_rate: f64, ) -> u64

Estimates output bitrate in bits/s from a floating-point QP value.

• qp – Quantization parameter.
• width – Frame width in pixels.
• height – Frame height in pixels.
• frame_rate – Frames per second.

pub fn estimate_from_vmaf( &self, vmaf: f64, width: u32, height: u32, frame_rate: f64, ) -> u64

Estimates bitrate from a target VMAF score (0–100).

Linearly maps VMAF → effective QP, then delegates to estimate_from_qp. VMAF 100 ≈ QP 0 (lossless), VMAF 0 ≈ QP 51 (worst).

pub fn crf_for_target_bitrate( &self, target_bps: u64, width: u32, height: u32, frame_rate: f64, ) -> Option<u8>

Infers the CRF value that would be required to hit a target bitrate.

Returns None when the target cannot be reached with valid QP values (0–63).

pub fn estimate_file_size( &self, crf: u8, width: u32, height: u32, frame_rate: f64, duration_secs: f64, ) -> u64

Returns an estimated size in bytes for encoding duration_secs of video.

Trait Implementations

impl Debug for BitrateEstimator

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for BitrateEstimator

fn default() -> Self

Returns the “default value” for a type.